Deployment Guide Red Hat Enterprise Linux 6 Barbora Ancincova

User Manual:

Open the PDF directly: View PDF .
Page Count: 830

Download
Open PDF In Browser	View PDF

Red Hat Enterprise Linux 6
Deployment Guide

Deployment, Configuration and Administration of Red Hat Enterprise Linux
6

Barbora Ančincová
Douglas Silas
Eva Kopalová
Tomáš Čapek
Petr Bokoč
Milan Navrátil
John Ha
Don Domingo

Eva Majoršinová
Martin Prpič
Peter Ondrejka
Petr Kovář
Jiří Herrmann
Robert Krátký
David O'Brien

Jaromír Hradílek
Stephen Wadeley
Ella Lackey
Miroslav Svoboda
Jana Švárová
Florian Nadge
Michael Hideo

Red Hat Enterprise Linux 6 Deployment Guide

Deployment, Configuration and Administration of Red Hat Enterprise Linux
6
Barbo ra Ančinco vá
Red Hat Custo mer Co ntent Services
bancinco @redhat.co m
Eva Majo ršino vá
Red Hat Custo mer Co ntent Services
emajo rsi@redhat.co m
Jaro mír Hradílek
Red Hat Custo mer Co ntent Services
jhradilek@redhat.co m
Do uglas Silas
Red Hat Custo mer Co ntent Services
silas@redhat.co m
Martin Prpič
Red Hat Custo mer Co ntent Services
mprpic@redhat.co m
Stephen Wadeley
Red Hat Custo mer Co ntent Services
swadeley@redhat.co m
Eva Ko palo vá
Red Hat Custo mer Co ntent Services
eko palo va@redhat.co m
Peter Ondrejka
Red Hat Custo mer Co ntent Services
po ndrejk@redhat.co m
Ella Lackey
Red Hat Custo mer Co ntent Services
dlackey@redhat.co m
To máš Čapek
Red Hat Custo mer Co ntent Services
tcapek@redhat.co m
Petr Ko vář
Red Hat Custo mer Co ntent Services
pko var@redhat.co m
Miro slav Svo bo da
Red Hat Custo mer Co ntent Services
msvo bo da@redhat.co m
Petr Bo ko č

Red Hat Custo mer Co ntent Services
pbo ko c@redhat.co m
Jiří Herrmann
Red Hat Custo mer Co ntent Services
jherrmann@redhat.co m
Jana Šváro vá
Red Hat Custo mer Co ntent Services
jsvaro va@redhat.co m
Milan Navrátil
Red Hat Custo mer Co ntent Services
mnavrati@redhat.co m
Ro bert Krátký
Red Hat Custo mer Co ntent Services
rkratky@redhat.co m
Flo rian Nadge
Red Hat Custo mer Co ntent Services
Jo hn Ha
Red Hat Custo mer Co ntent Services
David O'Brien
Red Hat Custo mer Co ntent Services
Michael Hideo
Red Hat Custo mer Co ntent Services
Do n Do mingo
Red Hat Custo mer Co ntent Services

Legal Notice
Co pyright © 20 10 –20 15 Red Hat, Inc.
This do cument is licensed by Red Hat under the Creative Co mmo ns Attributio n-ShareAlike 3.0
Unpo rted License. If yo u distribute this do cument, o r a mo dified versio n o f it, yo u must pro vide
attributio n to Red Hat, Inc. and pro vide a link to the o riginal. If the do cument is mo dified, all Red
Hat trademarks must be remo ved.
Red Hat, as the licenso r o f this do cument, waives the right to enfo rce, and agrees no t to assert,
Sectio n 4 d o f CC-BY-SA to the fullest extent permitted by applicable law.
Red Hat, Red Hat Enterprise Linux, the Shado wman lo go , JBo ss, MetaMatrix, Fedo ra, the Infinity
Lo go , and RHCE are trademarks o f Red Hat, Inc., registered in the United States and o ther
co untries.
Linux ® is the registered trademark o f Linus To rvalds in the United States and o ther co untries.
Java ® is a registered trademark o f Oracle and/o r its affiliates.
XFS ® is a trademark o f Silico n Graphics Internatio nal Co rp. o r its subsidiaries in the United
States and/o r o ther co untries.
MySQL ® is a registered trademark o f MySQL AB in the United States, the Euro pean Unio n and
o ther co untries.
No de.js ® is an o fficial trademark o f Jo yent. Red Hat So ftware Co llectio ns is no t fo rmally
related to o r endo rsed by the o fficial Jo yent No de.js o pen so urce o r co mmercial pro ject.
The OpenStack ® Wo rd Mark and OpenStack Lo go are either registered trademarks/service
marks o r trademarks/service marks o f the OpenStack Fo undatio n, in the United States and o ther
co untries and are used with the OpenStack Fo undatio n's permissio n. We are no t affiliated with,
endo rsed o r spo nso red by the OpenStack Fo undatio n, o r the OpenStack co mmunity.
All o ther trademarks are the pro perty o f their respective o wners.

Abstract
The Deplo yment Guide do cuments relevant info rmatio n regarding the deplo yment,
co nfiguratio n and administratio n o f Red Hat Enterprise Linux 6 . It is o riented to wards system
administrato rs with a basic understanding o f the system.

T able of Cont ent s

T able of Contents
. .art
⁠P
. . .I.. Basic
. . . . . Syst
. . . . em
. . . Configurat
. . . . . . . . . .ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 6. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .1. .. Keyboard
. . . . . . . . .Configurat
. . . . . . . . . ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 7. . . . . . . . . .
⁠1.1. Chang ing the Keyb o ard Layo ut
17
⁠1.2. Ad d ing the Keyb o ard Layo ut Ind ic ato r
19
⁠1.3. Setting Up a Typ ing Break
21
. .hapt
⁠C
. . . .er
. .2. .. Dat
. . . e. .and
. . . .T.ime
. . . Configurat
. . . . . . . . . .ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. 3. . . . . . . . . .
⁠2 .1. Date/Time Pro p erties To o l
23
⁠2 .1.1. Date and Time Pro p erties
23
⁠2 .1.2. Netwo rk Time Pro to c o l Pro p erties
24
⁠2 .1.3. Time Zo ne Pro p erties
25
⁠2 .2. Co mmand Line Co nfig uratio n
26
⁠2 .2.1. Date and Time Setup
27
⁠2 .2.2. Netwo rk Time Pro to c o l Setup
27
. .hapt
⁠C
. . . .er
. .3.
. .Managing
. . . . . . . . . Users
. . . . . .and
...G
. .roups
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
...........
⁠3 .1. What Us ers and G ro up s Are
30
⁠3 .2. Manag ing Us ers via the Us er Manag er Ap p lic atio n
⁠3 .2.1. Viewing Us ers
⁠3 .2.2. Ad d ing a New Us er
⁠3 .2.3. Mo d ifying Us er Pro p erties

31
31
32
33

⁠3 .3. Manag ing G ro up s via the Us er Manag er Ap p lic atio n
⁠3 .3.1. Viewing G ro up s
⁠3 .3.2. Ad d ing a New G ro up

34
34
35

⁠ .3.3. Mo d ifying G ro up Pro p erties
3
⁠3 .4. Manag ing Us ers via Co mmand -Line To o ls

36
36

⁠3 .4.1. Creating Us ers
⁠3 .4.2. Attac hing New Us ers to G ro up s

36
39

⁠3 .4.3. Up d ating Us ers ' Authentic atio n
⁠3 .4.4. Mo d ifying Us er Setting s

40
41

⁠3 .4.5. Deleting Us ers
⁠3 .4.6 . Dis p laying Co mp rehens ive Us er Info rmatio n

42
42

⁠3 .5. Manag ing G ro up s via Co mmand -Line To o ls
⁠3 .5.1. Creating G ro up s
⁠3 .5.2. Attac hing Us ers to G ro up s
⁠3 .5.3. Up d ating G ro up Authentic atio n

45
45
46
47

⁠3 .5.4. Mo d ifying G ro up Setting s
⁠3 .5.5. Deleting G ro up s
⁠3 .6 . Ad d itio nal Res o urc es
⁠3 .6 .1. Ins talled Do c umentatio n

47
48
48
48

. .hapt
⁠C
. . . .er
. .4. .. G
. .aining
. . . . . .Privileges
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
...........
⁠4 .1. The s u Co mmand
⁠4 .2. The s ud o Co mmand
⁠4 .3. Ad d itio nal Res o urc es
⁠Ins talled Do c umentatio n
⁠O nline Do c umentatio n

50
51
52
52
53

. .art
⁠P
. . .II.. .Subscript
. . . . . . . . ion
. . . .and
. . . Support
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
...........
. .hapt
⁠C
. . . .er
. .5.
. .Regist
. . . . . .ering
. . . . .t .he
. . Syst
. . . . em
. . . and
. . . . Managing
. . . . . . . . . Subscript
. . . . . . . . .ions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
...........
⁠5 .1. Reg is tering the Sys tem and Attac hing Sub s c rip tio ns
55
⁠5 .2. Manag ing So ftware Rep o s ito ries
56

1

Deployment G uide
⁠5 .2. Manag ing So ftware Rep o s ito ries
⁠5 .3. Remo ving Sub s c rip tio ns
⁠5 .4. Ad d itio nal Res o urc es
⁠Ins talled Do c umentatio n
⁠R elated Bo o ks
⁠O nline Res o urc es
⁠S ee Als o

56
57
57
57
57
58
58

. .hapt
⁠C
. . . .er
. .6. .. Accessing
. . . . . . . . . .Support
. . . . . . . Using
. . . . . . t.he
. . .Red
. . . .Hat
. . . Support
. . . . . . . . T. ool
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
...........
⁠6 .1. Ins talling the Red Hat Sup p o rt To o l
⁠6 .2. Reg is tering the Red Hat Sup p o rt To o l Us ing the Co mmand Line
⁠6 .3. Us ing the Red Hat Sup p o rt To o l in Interac tive Shell Mo d e
⁠6 .4. Co nfig uring the Red Hat Sup p o rt To o l

59
59
59
59

⁠6 .4.1. Saving Setting s to the Co nfig uratio n Files
⁠6 .5. O p ening and Up d ating Sup p o rt Cas es Us ing Interac tive Mo d e
⁠6 .6 . Viewing Sup p o rt Cas es o n the Co mmand Line
⁠6 .7. Ad d itio nal Res o urc es

60
61
63
63

. .art
⁠P
. . .III.
. . Inst
. . . .alling
. . . . .and
. . . .Managing
. . . . . . . . .Soft
. . . .ware
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. 4. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .7. .. Yum
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. 5. . . . . . . . . .
⁠7 .1. Chec king Fo r and Up d ating Pac kag es
⁠7 .1.1. Chec king Fo r Up d ates
⁠7 .1.2. Up d ating Pac kag es

66

⁠7 .1.3. Pres erving Co nfig uratio n File Chang es
⁠7 .1.4. Up g rad ing the Sys tem O ff-line with ISO and Yum

68
68

⁠7 .2. Pac kag es and Pac kag e G ro up s
⁠7 .2.1. Searc hing Pac kag es

70
70

⁠7 .2.2. Lis ting Pac kag es

71

⁠7 .2.3. Dis p laying Pac kag e Info rmatio n
⁠7 .2.4. Ins talling Pac kag es

73
74

⁠ .2.5. Remo ving Pac kag es
7
⁠7 .3. Wo rking with Trans ac tio n His to ry

76
77

⁠7 .3.1. Lis ting Trans ac tio ns

77

⁠7 .3.2. Examining Trans ac tio ns
⁠7 .3.3. Reverting and Rep eating Trans ac tio ns

80
81

⁠7 .3.4. Co mp leting Trans ac tio ns

82

⁠7 .3.5. Starting New Trans ac tio n His to ry
⁠7 .4. Co nfig uring Yum and Yum Rep o s ito ries

82
82

⁠7 .4.1. Setting [main] O p tio ns
⁠7 .4.2. Setting [rep o s ito ry] O p tio ns

83
86

⁠7 .4.3. Us ing Yum Variab les

88

⁠7 .4.4. Viewing the Current Co nfig uratio n
⁠7 .4.5. Ad d ing , Enab ling , and Dis ab ling a Yum Rep o s ito ry

89
89

⁠7 .4.6 . Creating a Yum Rep o s ito ry
⁠7 .4.7. Wo rking with Yum Cac he

91
92

⁠7 .4.8 . Ad d ing the O p tio nal and Sup p lementary Rep o s ito ries

93

⁠7 .5. Yum Plug -ins
⁠7 .5.1. Enab ling , Co nfig uring , and Dis ab ling Yum Plug -ins
⁠7 .5.2. Ins talling Ad d itio nal Yum Plug -ins

2

65
65

94
94
95

⁠ .5.3. Plug -in Des c rip tio ns
7
⁠7 .6 . Ad d itio nal Res o urc es

95
98

⁠Ins talled Do c umentatio n
⁠O nline Res o urc es

99
99

⁠S ee Als o

99

T able of Cont ent s
⁠S ee Als o

99

. .hapt
⁠C
. . . .er
. .8. .. PackageKit
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.0. 0. . . . . . . . . .
⁠8 .1. Up d ating Pac kag es with So ftware Up d ate
⁠ Setting the Up d ate-Chec king Interval

10 0
10 1

⁠8 .2. Us ing Ad d /Remo ve So ftware
⁠8 .2.1. Refres hing So ftware So urc es (Yum Rep o s ito ries )

10 2
10 2

⁠8 .2.2. Find ing Pac kag es with Filters

10 3

⁠8 .2.3. Ins talling and Remo ving Pac kag es (and Dep end enc ies )
⁠8 .2.4. Ins talling and Remo ving Pac kag e G ro up s

10 5
10 6

⁠ .2.5. Viewing the Trans ac tio n Lo g
8
⁠8 .3. Pac kag eKit Arc hitec ture

10 7
10 8

⁠8 .4. Ad d itio nal Res o urc es

10 9

⁠Ins talled Do c umentatio n
⁠O nline Do c umentatio n

10 9
110

⁠S ee Als o

110

. .art
⁠P
. . .IV.
. . Net
. . . working
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.1. 1. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .9. .. Net
. . . workManager
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.1. 2. . . . . . . . . .
⁠9 .1. The Netwo rkManag er Daemo n
⁠9 .2. Interac ting with Netwo rkManag er

112
112

⁠9 .2.1. Co nnec ting to a Netwo rk
⁠9 .2.2. Co nfig uring New and Ed iting Exis ting Co nnec tio ns

113
114

⁠9 .2.3. Co nnec ting to a Netwo rk Auto matic ally

115

⁠9 .2.4. Us er and Sys tem Co nnec tio ns
⁠9 .3. Es tab lis hing Co nnec tio ns

116
117

⁠9 .3.1. Es tab lis hing a Wired (Ethernet) Co nnec tio n
⁠9 .3.2. Es tab lis hing a Wireles s Co nnec tio n

117
121

⁠9 .3.3. Es tab lis hing a Mo b ile Bro ad b and Co nnec tio n

126

⁠9 .3.4. Es tab lis hing a VPN Co nnec tio n
⁠9 .3.5. Es tab lis hing a DSL Co nnec tio n

129
133

⁠9 .3.6 . Es tab lis hing a Bo nd Co nnec tio n

134

⁠9 .3.7. Es tab lis hing a VLAN Co nnec tio n
⁠9 .3.8 . Es tab lis hing an IP-o ver-InfiniBand (IPo IB) Co nnec tio n

139
141

⁠9 .3.9 . Co nfig uring Co nnec tio n Setting s

143

. .hapt
⁠C
. . . .er
. .1. 0. .. Net
. . . work
. . . . .Int
. . erfaces
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 52
...........
⁠10 .1. Netwo rk Co nfig uratio n Files
152
⁠10 .2. Interfac e Co nfig uratio n Files

153

⁠10 .2.1. Ethernet Interfac es

153

⁠10 .2.2. Sp ec ific ifc fg O p tio ns fo r Linux o n Sys tem z
⁠10 .2.3. Req uired ifc fg O p tio ns fo r Linux o n Sys tem z

158
159

⁠10 .2.4. Channel Bo nd ing Interfac es

159

⁠10 .2.5. Co nfig uring a VLAN o ver a Bo nd

16 3

⁠10 .2.6 . Netwo rk Brid g e

16 9

⁠10 .2.7. Setting Up 8 0 2.1Q VLAN Tag g ing
⁠10 .2.8 . Alias and Clo ne Files

173
174

⁠10 .2.9 . Dialup Interfac es

175

⁠10 .2.10 . O ther Interfac es

177

⁠10 .3. Interfac e Co ntro l Sc rip ts

178

⁠10 .4. Static Ro utes and the Default G ateway
⁠ Co nfig uring Static Ro utes Us ing the Co mmand Line

18 0
18 0

⁠ Co nfig uring The Default G ateway
⁠10 .5. Co nfig uring Static Ro utes in ifc fg files

18 0
18 1

3

Deployment G uide
⁠10 .5. Co nfig uring Static Ro utes in ifc fg files

18 1

⁠10 .5.1. Static Ro utes Us ing the IP Co mmand Arg uments Fo rmat

18 1

⁠10 .5.2. Netwo rk/Netmas k Direc tives Fo rmat

18 2

⁠10 .6 . Co nfig uring IPv6 To keniz ed Interfac e Id entifiers
⁠10 .7. Netwo rk Func tio n Files

18 3
18 4

⁠10 .8 . Ethto o l

18 4

⁠10 .9 . Ad d itio nal Res o urc es

19 0

⁠Ins talled Do c umentatio n

19 1

⁠O nline Res o urc es
⁠S ee Als o

19 1
19 1

. .art
⁠P
. . .V.. .Infrast
. . . . . .ruct
. . . ure
. . . .Services
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.9. 2. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .1. 1. .. Services
. . . . . . . . and
. . . .Daemons
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 9. 3. . . . . . . . . .
⁠11.1. Co nfig uring the Default Runlevel

19 3

⁠11.2. Co nfig uring the Servic es

19 4

⁠11.2.1. Us ing the Servic e Co nfig uratio n Utility

19 4

⁠11.2.2. Us ing the nts ys v Utility
⁠11.2.3. Us ing the c hkc o nfig Utility

19 6
19 7

⁠11.3. Running Servic es

19 9

⁠11.3.1. Determining the Servic e Status

19 9

⁠11.3.2. Starting a Servic e

20 0

⁠11.3.3. Sto p p ing a Servic e
⁠11.3.4. Res tarting a Servic e

20 0
20 0

⁠11.4. Ad d itio nal Res o urc es

20 1

⁠11.4.1. Ins talled Do c umentatio n

20 1

⁠11.4.2. Related Bo o ks

20 1

. .hapt
⁠C
. . . .er
. .1. 2. .. Configuring
. . . . . . . . . . . Aut
. . . hent
. . . . .icat
. . .ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.0. 2. . . . . . . . . .
⁠12.1. Co nfig uring Sys tem Authentic atio n
⁠12.1.1. Launc hing the Authentic atio n Co nfig uratio n To o l UI
⁠12.1.2. Selec ting the Id entity Sto re fo r Authentic atio n

20 3

⁠12.1.3. Co nfig uring Alternative Authentic atio n Features

210

⁠12.1.4. Co nfig uring Authentic atio n fro m the Co mmand Line

212

⁠12.1.5. Us ing Cus to m Ho me Direc to ries

215

⁠12.2. Us ing and Cac hing Cred entials with SSSD
⁠12.2.1. Ab o ut SSSD

4

20 2
20 2

216
216

⁠12.2.2. Setting up the s s s d .c o nf File

217

⁠12.2.3. Starting and Sto p p ing SSSD

219

⁠12.2.4. SSSD and Sys tem Servic es

219

⁠12.2.5. Co nfig uring Servic es : NSS
⁠12.2.6 . Co nfig uring Servic es : PAM

220
223

⁠12.2.7. Co nfig uring Servic es : auto fs

225

⁠12.2.8 . Co nfig uring Servic es : s ud o

228

⁠12.2.9 . Co nfig uring Servic es : O p enSSH and Cac hed Keys

231

⁠12.2.10 . SSSD and Id entity Pro vid ers (Do mains )
⁠12.2.11. Creating Do mains : LDAP

234
237

⁠12.2.12. Creating Do mains : Id entity Manag ement (Id M)

242

⁠12.2.13. Creating Do mains : Ac tive Direc to ry

244

⁠12.2.14. Co nfig uring Do mains : Ac tive Direc to ry as an LDAP Pro vid er (Alternative)

250

⁠12.2.15. Do main O p tio ns : Setting Us ername Fo rmats
⁠12.2.16 . Do main O p tio ns : Enab ling O ffline Authentic atio n

254
255

⁠12.2.17. Do main O p tio ns : Setting Pas s wo rd Exp iratio ns

257

⁠12.2.18 . Do main O p tio ns : Us ing DNS Servic e Dis c o very

258

⁠12.2.19 . Do main O p tio ns : Us ing IP Ad d res s es in Certific ate Sub jec t Names (LDAP O nly)

26 0

T able of Cont ent s
⁠12.2.19 . Do main O p tio ns : Us ing IP Ad d res s es in Certific ate Sub jec t Names (LDAP O nly)

26 0

⁠12.2.20 . Creating Do mains : Pro xy
⁠12.2.21. Creating Do mains : Kerb ero s Authentic atio n

26 1
26 3

⁠12.2.22. Creating Do mains : Ac c es s Co ntro l

26 6

⁠12.2.23. Creating Do mains : Primary Server and Bac kup Servers

26 8

⁠12.2.24. Ins talling SSSD Utilities

26 8

⁠12.2.25. SSSD and UID and G ID Numb ers
⁠12.2.26 . Creating Lo c al Sys tem Us ers

26 9
26 9

⁠12.2.27. Seed ing Us ers into the SSSD Cac he During Kic ks tart

270

⁠12.2.28 . Manag ing the SSSD Cac he

271

⁠12.2.29 . Do wng rad ing SSSD

272

⁠12.2.30 . Us ing NSCD with SSSD
⁠12.2.31. Tro ub les ho o ting SSSD

273
273

. .hapt
⁠C
. . . .er
. .1. 3.
. .O
. .penSSH
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.8. 1. . . . . . . . . .
⁠13.1. The SSH Pro to c o l
28 1
⁠13.1.1. Why Us e SSH?

28 1

⁠13.1.2. Main Features

28 1

⁠13.1.3. Pro to c o l Vers io ns

28 2

⁠13.1.4. Event Seq uenc e o f an SSH Co nnec tio n
⁠13.2. Co nfig uring O p enSSH
⁠13.2.1. Co nfig uratio n Files

28 2
28 4
28 4

⁠13.2.2. Starting an O p enSSH Server

28 5

⁠13.2.3. Req uiring SSH fo r Remo te Co nnec tio ns

28 6

⁠13.2.4. Us ing Key-Bas ed Authentic atio n
⁠13.3. Us ing O p enSSH Certific ate Authentic atio n

28 6
29 2

⁠13.3.1. Intro d uc tio n to SSH Certific ates

29 2

⁠13.3.2. Sup p o rt fo r SSH Certific ates

29 3

⁠13.3.3. Creating SSH CA Certific ate Sig ning Keys

29 3

⁠13.3.4. Dis trib uting and Trus ting SSH CA Pub lic Keys
⁠13.3.5. Creating SSH Certific ates

29 5
29 7

⁠13.3.6 . Sig ning an SSH Certific ate Us ing a PKCS#11 To ken

30 1

⁠13.3.7. Viewing an SSH CA Certific ate

30 2

⁠13.3.8 . Revo king an SSH CA Certific ate

30 2

⁠13.4. O p enSSH Clients
⁠13.4.1. Us ing the s s h Utility

30 3
30 3

⁠13.4.2. Us ing the s c p Utility

30 4

⁠13.4.3. Us ing the s ftp Utility

30 5

⁠13.5. Mo re Than a Sec ure Shell

30 6

⁠13.5.1. X11 Fo rward ing
⁠13.5.2. Po rt Fo rward ing

30 6
30 6

⁠13.6 . Ad d itio nal Res o urc es

30 7

⁠13.6 .1. Ins talled Do c umentatio n

30 7

⁠13.6 .2. Us eful Web s ites

30 8

. .hapt
⁠C
. . . .er
. .1. 4. .. T
. .igerVNC
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .30
. . 9. . . . . . . . . .
⁠14.1. VNC Server
⁠14.1.1. Ins talling VNC Server
⁠14.1.2. Co nfig uring VNC Server

30 9
30 9
30 9

⁠14.1.3. Starting VNC Server

311

⁠14.1.4. Terminating a VNC Ses s io n

312

⁠14.2. Sharing an Exis ting Des kto p

312

⁠14.3. Us ing a VNC Viewer
⁠14.3.1. Ins talling the VNC Viewer

313
313

5

Deployment G uide
⁠14.3.2. Co nnec ting to a VNC Server

313

⁠14.3.3. Co nnec ting to VNC Server Us ing SSH

316

⁠14.4. Ad d itio nal Res o urc es
⁠Ins talled Do c umentatio n

316
316

. .art
⁠P
. . .VI.
. . Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
. . 7. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .1. 5.
. . DHCP
. . . . . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
. . 8. . . . . . . . . .
⁠15.1. Why Us e DHCP?
318
⁠15.2. Co nfig uring a DHCP Server

318

⁠15.2.1. Co nfig uratio n File

318

⁠15.2.2. Leas e Datab as e

321

⁠15.2.3. Starting and Sto p p ing the Server
⁠15.2.4. DHCP Relay Ag ent

322
323

⁠15.3. Co nfig uring a DHCP Client

323

⁠15.4. Co nfig uring a Multiho med DHCP Server

324

⁠15.4.1. Ho s t Co nfig uratio n

325

⁠15.5. DHCP fo r IPv6 (DHCPv6 )
⁠15.6 . Ad d itio nal Res o urc es

327
327

⁠15.6 .1. Ins talled Do c umentatio n

327

. .hapt
⁠C
. . . .er
. .1. 6. .. DNS
. . . . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32
. . 9. . . . . . . . . .
⁠16 .1. Intro d uc tio n to DNS
329
⁠16 .1.1. Names erver Zo nes
329
⁠16 .1.2. Names erver Typ es
⁠16 .1.3. BIND as a Names erver
⁠16 .2. BIND
⁠16 .2.1. Co nfig uring the named Servic e
⁠16 .2.2. Ed iting Zo ne Files

329
330
330
330
338

⁠16 .2.3. Us ing the rnd c Utility
⁠16 .2.4. Us ing the d ig Utility

345
348

⁠16 .2.5. Ad vanc ed Features o f BIND
⁠16 .2.6 . Co mmo n Mis takes to Avo id
⁠16 .2.7. Ad d itio nal Res o urc es

350
351
352

. .hapt
⁠C
. . . .er
. .1. 7. .. Web
. . . . Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
............
⁠17.1. The Ap ac he HTTP Server
⁠17.1.1. New Features

354
354

⁠17.1.2. No tab le Chang es
⁠17.1.3. Up d ating the Co nfig uratio n
⁠17.1.4. Running the http d Servic e

354
354
355

⁠17.1.5. Ed iting the Co nfig uratio n Files
⁠17.1.6 . Wo rking with Mo d ules

356
38 7

⁠17.1.7. Setting Up Virtual Ho s ts
⁠17.1.8 . Setting Up an SSL Server

38 8
38 9

⁠17.1.9 . Enab ling the mo d _s s l Mo d ule
⁠17.1.10 . Enab ling the mo d _ns s Mo d ule
⁠17.1.11. Us ing an Exis ting Key and Certific ate

39 0
39 3
39 8

⁠17.1.12. G enerating a New Key and Certific ate
⁠17.1.13. Co nfig ure the Firewall fo r HTTP and HTTPS Us ing the Co mmand Line

39 9
40 5

⁠17.1.14. Ad d itio nal Res o urc es

40 6

. .hapt
⁠C
. . . .er
. .1. 8. .. Mail
. . . . Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.0. 8. . . . . . . . . .
⁠18 .1. Email Pro to c o ls
40 8
⁠18 .1.1. Mail Trans p o rt Pro to c o ls
40 8

6

T able of Cont ent s
⁠18 .1.1. Mail Trans p o rt Pro to c o ls

40 8

⁠ 8 .1.2. Mail Ac c es s Pro to c o ls
1
⁠18 .2. Email Pro g ram Clas s ific atio ns

40 8
411

⁠18 .2.1. Mail Trans p o rt Ag ent
⁠18 .2.2. Mail Delivery Ag ent
⁠18 .2.3. Mail Us er Ag ent
⁠18 .3. Mail Trans p o rt Ag ents
⁠18 .3.1. Po s tfix
⁠18 .3.2. Send mail
⁠18 .3.3. Fetc hmail
⁠18 .3.4. Mail Trans p o rt Ag ent (MTA) Co nfig uratio n

411
411
412
412
412
414
419
423

⁠18 .4. Mail Delivery Ag ents
⁠18 .4.1. Pro c mail Co nfig uratio n

423
424

⁠18 .4.2. Pro c mail Rec ip es
⁠18 .5. Mail Us er Ag ents
⁠18 .5.1. Sec uring Co mmunic atio n

425
430
430

⁠18 .6 . Ad d itio nal Res o urc es
⁠18 .6 .1. Ins talled Do c umentatio n

431
431

⁠18 .6 .2. O nline Do c umentatio n
⁠18 .6 .3. Related Bo o ks

432
432

. .hapt
⁠C
. . . .er
. .1. 9. .. Direct
. . . . . .ory
. . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. 33
...........
⁠19 .1. O p enLDAP
433
⁠19 .1.1. Intro d uc tio n to LDAP
⁠19 .1.2. Ins talling the O p enLDAP Suite

433
435

⁠19 .1.3. Co nfig uring an O p enLDAP Server
⁠19 .1.4. Running an O p enLDAP Server
⁠19 .1.5. Co nfig uring a Sys tem to Authentic ate Us ing O p enLDAP

438
442
443

⁠19 .1.6 . Ad d itio nal Res o urc es

444

. .hapt
⁠C
. . . .er
. .2. 0. .. File
. . . .and
. . . .Print
. . . . Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.4. 7. . . . . . . . . .
⁠2 0 .1. Samb a
447
⁠2 0 .1.1. Intro d uc tio n to Samb a
⁠2 0 .1.2. Samb a Daemo ns and Related Servic es
⁠2 0 .1.3. Co nnec ting to a Samb a Share

447
447
448

⁠2 0 .1.4. Co nfig uring a Samb a Server
⁠2 0 .1.5. Starting and Sto p p ing Samb a

451
452

⁠2 0 .1.6 . Samb a Server Typ es and the s mb .c o nf File
⁠2 0 .1.7. Samb a Sec urity Mo d es
⁠2 0 .1.8 . Samb a Ac c o unt Info rmatio n Datab as es

453
46 1
46 3

⁠2 0 .1.9 . Samb a Netwo rk Bro ws ing
⁠2 0 .1.10 . Samb a with CUPS Printing Sup p o rt

46 4
46 5

⁠2 0 .1.11. Samb a Dis trib utio n Pro g rams
⁠2 0 .1.12. Ad d itio nal Res o urc es
⁠2 0 .2. FTP
⁠2 0 .2.1. The File Trans fer Pro to c o l
⁠2 0 .2.2. The vs ftp d Server
⁠2 0 .2.3. Ad d itio nal Res o urc es
⁠2 0 .3. Printer Co nfig uratio n
⁠2 0 .3.1. Starting the Printer Co nfig uratio n To o l

46 6
471
472
472
473
48 6
48 7
48 8

⁠2 0 .3.2. Starting Printer Setup
⁠2 0 .3.3. Ad d ing a Lo c al Printer

48 8
48 9

⁠2 0 .3.4. Ad d ing an Ap p So c ket/HP JetDirec t p rinter
⁠2 0 .3.5. Ad d ing an IPP Printer

49 0
49 1

⁠2 0 .3.6 . Ad d ing an LPD/LPR Ho s t o r Printer

49 2

7

Deployment G uide
⁠2 0 .3.6 . Ad d ing an LPD/LPR Ho s t o r Printer
⁠2 0 .3.7. Ad d ing a Samb a (SMB) p rinter
⁠2 0 .3.8 . Selec ting the Printer Mo d el and Finis hing

49 2
49 3
49 5

⁠2 0 .3.9 . Printing a Tes t Pag e
⁠2 0 .3.10 . Mo d ifying Exis ting Printers

49 8
49 8

⁠2 0 .3.11. Ad d itio nal Res o urc es

50 4

. .hapt
⁠C
. . . .er
. .2. 1. .. Configuring
. . . . . . . . . . . NT
. . .P. .Using
. . . . . nt
. . pd
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50
. . 6. . . . . . . . . .
⁠2 1.1. Intro d uc tio n to NTP
50 6
⁠2 1.2. NTP Strata
50 6
⁠2 1.3. Und ers tand ing NTP
⁠2 1.4. Und ers tand ing the Drift File

50 7
50 8

⁠2 1.5. UTC, Timez o nes , and DST
⁠2 1.6 . Authentic atio n O p tio ns fo r NTP
⁠2 1.7. Manag ing the Time o n Virtual Mac hines

50 8
50 8
50 9

⁠2 1.8 . Und ers tand ing Leap Sec o nd s
⁠2 1.9 . Und ers tand ing the ntp d Co nfig uratio n File

50 9
50 9

⁠2 1.10 . Und ers tand ing the ntp d Sys c o nfig File
⁠2 1.11. Chec king if the NTP Daemo n is Ins talled
⁠2 1.12. Ins talling the NTP Daemo n (ntp d )

511
511
512

⁠2 1.13. Chec king the Status o f NTP
⁠2 1.14. Co nfig ure the Firewall to Allo w Inc o ming NTP Pac kets

512
513

⁠2 1.14.1. Co nfig ure the Firewall Us ing the G rap hic al To o l
⁠2 1.14.2. Co nfig ure the Firewall Us ing the Co mmand Line
⁠2 1.15. Co nfig ure ntp d ate Servers

513
513
514

⁠2 1.16 . Co nfig ure NTP
⁠2 1.16 .1. Co nfig ure Ac c es s Co ntro l to an NTP Servic e

515
515

⁠2 1.16 .2. Co nfig ure Rate Limiting Ac c es s to an NTP Servic e
⁠2 1.16 .3. Ad d ing a Peer Ad d res s
⁠2 1.16 .4. Ad d ing a Server Ad d res s

516
516
516

⁠2 1.16 .5. Ad d ing a Bro ad c as t o r Multic as t Server Ad d res s
⁠2 1.16 .6 . Ad d ing a Manyc as t Client Ad d res s

517
517

⁠2 1.16 .7. Ad d ing a Bro ad c as t Client Ad d res s
⁠2 1.16 .8 . Ad d ing a Manyc as t Server Ad d res s
⁠2 1.16 .9 . Ad d ing a Multic as t Client Ad d res s

517
518
518

⁠2 1.16 .10 . Co nfig uring the Burs t O p tio n
⁠2 1.16 .11. Co nfig uring the ib urs t O p tio n

518
518

⁠2 1.16 .12. Co nfig uring Symmetric Authentic atio n Us ing a Key
⁠2 1.16 .13. Co nfig uring the Po ll Interval
⁠2 1.16 .14. Co nfig uring Server Preferenc e

519
519
519

⁠2 1.16 .15. Co nfig uring the Time-to -Live fo r NTP Pac kets
⁠2 1.16 .16 . Co nfig uring the NTP Vers io n to Us e

520
520

⁠2 1.17. Co nfig uring the Hard ware Clo c k Up d ate
⁠2 1.18 . Co nfig uring Clo c k So urc es
⁠2 1.19 . Ad d itio nal Res o urc es
⁠2 1.19 .1. Ins talled Do c umentatio n
⁠2 1.19 .2. Us eful Web s ites

520
520
521
521
521

. .hapt
⁠C
. . . .er
. .2. 2. .. Configuring
. . . . . . . . . . . PT
. . .P. Using
. . . . . . pt
. . p4
. . l. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .52
. . 2. . . . . . . . . .

8

⁠2 2.1. Intro d uc tio n to PTP
⁠2 2.1.1. Und ers tand ing PTP
⁠2 2.1.2. Ad vantag es o f PTP

522
522
523

⁠2 2.2. Us ing PTP
⁠2 2.2.1. Chec king fo r Driver and Hard ware Sup p o rt

524
524

T able of Cont ent s
⁠2 2.2.1. Chec king fo r Driver and Hard ware Sup p o rt

524

⁠2 2.2.2. Ins talling PTP
⁠2 2.2.3. Starting p tp 4l
⁠2 2.3. Sp ec ifying a Co nfig uratio n File

524
525
526

⁠2 2.4. Us ing the PTP Manag ement Client
⁠2 2.5. Sync hro niz ing the Clo c ks

527
527

⁠2 2.6 . Verifying Time Sync hro niz atio n
⁠2 2.7. Serving PTP Time With NTP
⁠2 2.8 . Serving NTP Time With PTP

528
530
530

⁠2 2.9 . Imp ro ving Ac c urac y
⁠2 2.10 . Ad d itio nal Res o urc es

531
531

⁠2 2.10 .1. Ins talled Do c umentatio n
⁠2 2.10 .2. Us eful Web s ites

531
531

. .art
⁠P
. . .VII.
. . .Monit
. . . . .oring
. . . . . and
. . . .Aut
. . . omat
. . . . .ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
............
. .hapt
⁠C
. . . .er
. .2. 3.
. . Syst
. . . . em
. . . Monit
. . . . . oring
.....T
. .ools
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533
............
⁠2 3.1. Viewing Sys tem Pro c es s es
⁠2 3.1.1. Us ing the p s Co mmand
⁠2 3.1.2. Us ing the to p Co mmand

533
533
534

⁠2 3.1.3. Us ing the Sys tem Mo nito r To o l
⁠2 3.2. Viewing Memo ry Us ag e

535
537

⁠2 3.2.1. Us ing the free Co mmand
⁠2 3.2.2. Us ing the Sys tem Mo nito r To o l
⁠2 3.3. Viewing CPU Us ag e

537
537
538

⁠2 3.3.1. Us ing the Sys tem Mo nito r To o l
⁠2 3.4. Viewing Blo c k Devic es and File Sys tems

538
539

⁠2 3.4.1. Us ing the ls b lk Co mmand
⁠2 3.4.2. Us ing the b lkid Co mmand
⁠2 3.4.3. Us ing the find mnt Co mmand

539
540
541

⁠2 3.4.4. Us ing the d f Co mmand
⁠2 3.4.5. Us ing the d u Co mmand

542
543

⁠ 3.4.6 . Us ing the Sys tem Mo nito r To o l
2
⁠2 3.5. Viewing Hard ware Info rmatio n
⁠2 3.5.1. Us ing the ls p c i Co mmand

544
544
544

⁠2 3.5.2. Us ing the ls us b Co mmand
⁠2 3.5.3. Us ing the ls p c mc ia Co mmand

545
546

⁠ 3.5.4. Us ing the ls c p u Co mmand
2
⁠2 3.6 . Mo nito ring Perfo rmanc e with Net-SNMP
⁠2 3.6 .1. Ins talling Net-SNMP
⁠2 3.6 .2. Running the Net-SNMP Daemo n
⁠2 3.6 .3. Co nfig uring Net-SNMP
⁠2 3.6 .4. Retrieving Perfo rmanc e Data o ver SNMP
⁠2 3.6 .5. Extend ing Net-SNMP
⁠2 3.7. Ad d itio nal Res o urc es
⁠2 3.7.1. Ins talled Do c umentatio n

546
547
547
548
549
552
555
56 0
56 0

. .hapt
⁠C
. . . .er
. .2. 4. .. Viewing
. . . . . . . .and
. . . Managing
. . . . . . . . . .Log
. . . Files
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56
. . 1. . . . . . . . . .
⁠2 4.1. Ins talling rs ys lo g
56 1
⁠ 4.1.1. Up g rad ing to rs ys lo g vers io n 7
2
⁠2 4.2. Lo c ating Lo g Files
⁠2 4.3. Bas ic Co nfig uratio n o f Rs ys lo g

56 1
56 2
56 2

⁠2 4.3.1. Filters
⁠2 4.3.2. Ac tio ns

56 3
56 6

⁠2 4.3.3. Temp lates

571

9

Deployment G uide
⁠2 4.3.3. Temp lates
⁠2 4.3.4. G lo b al Direc tives
⁠2 4.3.5. Lo g Ro tatio n

571
574
575

⁠2 4.4. Us ing the New Co nfig uratio n Fo rmat
⁠2 4.4.1. Rules ets

576
577

⁠ 4.4.2. Co mp atib ility with s ys klo g d
2
⁠2 4.5. Wo rking with Q ueues in Rs ys lo g
⁠2 4.5.1. Defining Q ueues

578
578
58 0

⁠2 4.5.2. Creating a New Direc to ry fo r rs ys lo g Lo g Files
⁠2 4.5.3. Manag ing Q ueues

58 3
58 4

⁠ 4.5.4. Us ing the New Syntax fo r rs ys lo g q ueues
2
⁠2 4.6 . Co nfig uring rs ys lo g o n a Lo g g ing Server
⁠2 4.6 .1. Us ing The New Temp late Syntax o n a Lo g g ing Server

58 6
58 7
59 0

⁠2 4.7. Us ing Rs ys lo g Mo d ules
⁠2 4.7.1. Imp o rting Text Files

59 1
59 2

⁠2 4.7.2. Exp o rting Mes s ag es to a Datab as e
⁠2 4.7.3. Enab ling Enc ryp ted Trans p o rt
⁠2 4.8 . Deb ug g ing Rs ys lo g

59 3
59 3
59 3

⁠2 4.9 . Manag ing Lo g Files in a G rap hic al Enviro nment
⁠2 4.9 .1. Viewing Lo g Files

59 4
59 4

⁠2 4.9 .2. Ad d ing a Lo g File
⁠2 4.9 .3. Mo nito ring Lo g Files
⁠2 4.10 . Ad d itio nal Res o urc es

59 7
59 8
59 9

⁠Ins talled Do c umentatio n
⁠O nline Do c umentatio n

59 9
59 9

⁠S ee Als o

59 9

. .hapt
⁠C
. . . .er
. .2. 5.
. . Aut
. . . omat
. . . . .ing
. . . Syst
. . . . em
. . . T. asks
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6.0. 1. . . . . . . . . .
⁠2 5.1. Cro n and Anac ro n
601
⁠2 5.1.1. Ins talling Cro n and Anac ro n
⁠2 5.1.2. Running the Cro nd Servic e
⁠2 5.1.3. Co nfig uring Anac ro n Jo b s

601
601
602

⁠2 5.1.4. Co nfig uring Cro n Jo b s
⁠2 5.1.5. Co ntro lling Ac c es s to Cro n

604
606

⁠2 5.1.6 . Blac k and White Lis ting o f Cro n Jo b s
⁠2 5.2. At and Batc h
⁠2 5.2.1. Ins talling At and Batc h

606
607
607

⁠2 5.2.2. Running the At Servic e
⁠2 5.2.3. Co nfig uring an At Jo b

607
608

⁠2 5.2.4. Co nfig uring a Batc h Jo b
⁠2 5.2.5. Viewing Pend ing Jo b s
⁠2 5.2.6 . Ad d itio nal Co mmand -Line O p tio ns

609
609
609

⁠ 5.2.7. Co ntro lling Ac c es s to At and Batc h
2
⁠2 5.3. Ad d itio nal Res o urc es

6 10
6 10

. .hapt
⁠C
. . . .er
. .2. 6. .. Aut
. . . omat
. . . . .ic
. .Bug
. . . . Report
. . . . . . ing
. . . .T.ool
. . . (ABRT
. . . . . .). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6.1. 1. . . . . . . . . .
⁠2 6 .1. Ins talling ABRT and Starting its Servic es
⁠2 6 .2. Us ing the G rap hic al Us er Interfac e
⁠2 6 .3. Us ing the Co mmand -Line Interfac e
⁠2 6 .3.1. Viewing Pro b lems
⁠2 6 .3.2. Rep o rting Pro b lems
⁠2 6 .3.3. Deleting Pro b lems
⁠2 6 .4. Co nfig uring ABRT
⁠2 6 .4.1. ABRT Events
⁠2 6 .4.2. Stand ard ABRT Ins tallatio n Sup p o rted Events

10

6 12
6 14
6 21
6 21
6 23
6 24
6 24
6 25
6 27

T able of Cont ent s
⁠2 6 .4.2. Stand ard ABRT Ins tallatio n Sup p o rted Events
⁠2 6 .4.3. Event Co nfig uratio n in ABRT G UI

6 27
6 27

⁠2 6 .4.4. ABRT Sp ec ific Co nfig uratio n
⁠2 6 .4.5. Co nfig uring ABRT to Detec t a Kernel Panic
⁠2 6 .4.6 . Auto matic Do wnlo ad s and Ins tallatio n o f Deb ug info Pac kag es

6 30
6 31
6 32

⁠2 6 .4.7. Co nfig uring Auto matic Rep o rting
⁠2 6 .4.8 . Up lo ad ing and Rep o rting Us ing a Pro xy Server

6 33
6 33

⁠2 6 .5. Co nfig uring Centraliz ed Cras h Co llec tio n
⁠2 6 .5.1. Co nfig uratio n Step s Req uired o n a Ded ic ated Sys tem
⁠2 6 .5.2. Co nfig uratio n Step s Req uired o n a Client Sys tem
⁠2 6 .5.3. Saving Pac kag e Info rmatio n
⁠2 6 .5.4. Tes ting ABRT' s Cras h Detec tio n

6 33
6 34
6 35
6 35
6 37

. .hapt
⁠C
. . . .er
. .2. 7. .. O
. .Profile
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. 38
...........
⁠2 7.1. O verview o f To o ls
⁠2 7.2. Co nfig uring O Pro file
⁠2 7.2.1. Sp ec ifying the Kernel
⁠2 7.2.2. Setting Events to Mo nito r
⁠2 7.2.3. Sep arating Kernel and Us er-s p ac e Pro files
⁠2 7.3. Starting and Sto p p ing O Pro file
⁠2 7.4. Saving Data
⁠2 7.5. Analyz ing the Data
⁠2 7.5.1. Us ing o p rep o rt
⁠2 7.5.2. Us ing o p rep o rt o n a Sing le Exec utab le

6 38
6 39
6 39
6 39
6 42
6 43
6 44
6 44
6 45
6 46

⁠2 7.5.3. G etting mo re d etailed o utp ut o n the mo d ules
⁠2 7.5.4. Us ing o p anno tate
⁠2 7.6 . Und ers tand ing /d ev/o p ro file/

6 47
6 48
6 49

⁠2 7.7. Examp le Us ag e
⁠2 7.8 . O Pro file Sup p o rt fo r Java

6 49
6 50

⁠ 7.8 .1. Pro filing Java Co d e
2
⁠2 7.9 . G rap hic al Interfac e
⁠2 7.10 . O Pro file and Sys temTap

6 50
6 50
6 53

⁠2 7.11. Ad d itio nal Res o urc es
⁠2 7.11.1. Ins talled Do c s

6 53
6 53

⁠2 7.11.2. Us eful Web s ites

6 53

. .art
⁠P
. . .VIII.
. . . Kernel,
. . . . . . .Module
. . . . . . .and
. . . .Driver
. . . . . Configurat
. . . . . . . . . .ion
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. 54
...........
. .hapt
⁠C
. . . .er
. .2. 8. .. Manually
. . . . . . . . .Upgrading
. . . . . . . . . t. he
. . .Kernel
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. 55
...........
⁠2 8 .1. O verview o f Kernel Pac kag es
6 55
⁠2 8 .2.
⁠2 8 .3.
⁠2 8 .4.
⁠2 8 .5.

Prep aring to Up g rad e
Do wnlo ad ing the Up g rad ed Kernel
Perfo rming the Up g rad e
Verifying the Initial RAM Dis k Imag e

⁠V erifying the Initial RAM Dis k Imag e and Kernel o n IBM eServer Sys tem i
⁠2 8 .6 . Verifying the Bo o t Lo ad er
⁠2 8 .6 .1. Co nfig uring the G RUB Bo o t Lo ad er
⁠2 8 .6 .2. Co nfig uring the Lo o p b ac k Devic e Limit
⁠2 8 .6 .3. Co nfig uring the O S/40 0 Bo o t Lo ad er
⁠2 8 .6 .4. Co nfig uring the YABO O T Bo o t Lo ad er

6 56
6 57
6 57
6 58
660
660
660
662
663
663

. .hapt
⁠C
. . . .er
. .2. 9. .. Working
. . . . . . . .wit
..h
. . Kernel
. . . . . . Modules
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. 6. 5. . . . . . . . . .
⁠2 9 .1. Lis ting Currently-Lo ad ed Mo d ules
665
⁠2 9 .2. Dis p laying Info rmatio n Ab o ut a Mo d ule
666
⁠2 9 .3. Lo ad ing a Mo d ule
668

11

Deployment G uide
⁠2 9 .3. Lo ad ing a Mo d ule
⁠2 9 .4. Unlo ad ing a Mo d ule

668
669

⁠2 9 .5. Blac klis ting a Mo d ule
⁠2 9 .6 . Setting Mo d ule Parameters
⁠2 9 .6 .1. Lo ad ing a Cus to miz ed Mo d ule - Temp o rary Chang es
⁠2 9 .6 .2. Lo ad ing a Cus to miz ed Mo d ule - Pers is tent Chang es
⁠2 9 .7. Pers is tent Mo d ule Lo ad ing

6 70
6 72
6 73
6 74
6 75

⁠2 9 .8 . Sp ec ific Kernel Mo d ule Cap ab ilities
⁠2 9 .8 .1. Us ing Channel Bo nd ing
⁠2 9 .9 . Ad d itio nal Res o urc es
⁠Ins talled Do c umentatio n
⁠Ins tallab le Do c umentatio n

6 75
6 75
681
681
682

⁠O nline Do c umentatio n

682

. .hapt
⁠C
. . . .er
. .30
. . .. T. he
. . . kdump
. . . . . . Crash
. . . . . .Recovery
. . . . . . . . .Service
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6. 8. 3. . . . . . . . . .
⁠3 0 .1. Ins talling the kd ump Servic e
683
⁠3 0 .2. Co nfig uring the kd ump Servic e
683
⁠3 0 .2.1. Co nfig uring kd ump at Firs t Bo o t
683
⁠3 0 .2.2. Us ing the Kernel Dump Co nfig uratio n Utility
684
⁠3 0 .2.3. Co nfig uring kd ump o n the Co mmand Line
⁠3 0 .2.4. Tes ting the Co nfig uratio n
⁠3 0 .3. Analyz ing the Co re Dump
⁠3 0 .3.1. Running the c ras h Utility
⁠3 0 .3.2. Dis p laying the Mes s ag e Buffer
⁠3 0 .3.3. Dis p laying a Bac ktrac e
⁠3 0 .3.4. Dis p laying a Pro c es s Status
⁠3 0 .3.5. Dis p laying Virtual Memo ry Info rmatio n
⁠3 0 .3.6 . Dis p laying O p en Files
⁠3 0 .3.7. Exiting the Utility
⁠3 0 .4. Ad d itio nal Res o urc es
⁠Ins talled Do c umentatio n
⁠U s eful Web s ites

689
694
694
695
696
697
697
698
699
699
699
699
70 0

. .art
⁠P
. . .IX.
. . Syst
. . . . em
. . . Recovery
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7.0. 1. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .31
. . .. Syst
. . . . em
. . . Recovery
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7.0. 2. . . . . . . . . .
⁠3 1.1. Res c ue Mo d e
70 2
⁠3 1.2. Sing le-Us er Mo d e
70 4
⁠3 1.3. Emerg enc y Mo d e
⁠3 1.4. Res o lving Pro b lems in Sys tem Rec o very Mo d es

70 4
70 5

. .ppendix
⁠A
. . . . . . . A.
. . Consist
. . . . . . . ent
. . . .Net
. . .work
. . . . Device
. . . . . . .Naming
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7.0. 8. . . . . . . . . .
⁠A .1. Affec ted Sys tems
70 8
⁠A .2. Sys tem Req uirements
70 8
⁠A .3. Enab ling and Dis ab ling the Feature
70 9
⁠A .4. No tes fo r Ad minis trato rs

70 9

. .ppendix
⁠A
. . . . . . . B.
. . .RPM
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7.1. 0. . . . . . . . . .
⁠B .1. RPM Des ig n G o als
711
⁠B .2. Us ing RPM
711
⁠B .2.1. Find ing RPM Pac kag es
712
⁠B .2.2. Ins talling and Up g rad ing
712
⁠B .2.3. Co nfig uratio n File Chang es
⁠B .2.4. Unins talling
⁠B .2.5. Fres hening
⁠B .2.6 . Q uerying

12

715
715
716
717

T able of Cont ent s
⁠B .2.6 . Q uerying
⁠B .2.7. Verifying
⁠B .3. Chec king a Pac kag e' s Sig nature
⁠B .3.1. Imp o rting Keys
⁠B .3.2. Verifying Sig nature o f Pac kag es
⁠B .4. Prac tic al and Co mmo n Examp les o f RPM Us ag e
⁠B .5. Ad d itio nal Res o urc es
⁠B .5.1. Ins talled Do c umentatio n
⁠B .5.2. Us eful Web s ites
⁠B .5.3. Related Bo o ks

717
717
718
719
719
719
721
721
722
722

. .ppendix
⁠A
. . . . . . . C.
..T
. .he
..X
. . Window
. . . . . . . Syst
. . . . em
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7. 2. 3. . . . . . . . . .
⁠C .1. The X Server
723
⁠C .2. Des kto p Enviro nments and Wind o w Manag ers
723
⁠C .2.1. Des kto p Enviro nments
⁠C .2.2. Wind o w Manag ers
⁠C .3. X Server Co nfig uratio n Files
⁠C .3.1. The Struc ture o f the Co nfig uratio n
⁠C .3.2. The xo rg .c o nf.d Direc to ry

724
724
725
725
726

⁠C .3.3. The xo rg .c o nf File
⁠C .4. Fo nts
⁠C .4.1. Ad d ing Fo nts to Fo ntc o nfig
⁠C .5. Runlevels and X
⁠C .5.1. Runlevel 3

726
733
733
734
734

⁠C .5.2. Runlevel 5
⁠C .6 . Ad d itio nal Res o urc es
⁠C .6 .1. Ins talled Do c umentatio n
⁠C .6 .2. Us eful Web s ites

735
736
736
736

. .ppendix
⁠A
. . . . . . . D.
. . .T.he
. . .sysconfig
. . . . . . . . .Direct
. . . . . ory
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7. 37
...........
⁠D .1. Files in the /etc /s ys c o nfig / Direc to ry
⁠D .1.1. /etc /s ys c o nfig /arp watc h
⁠D .1.2. /etc /s ys c o nfig /authc o nfig
⁠D .1.3. /etc /s ys c o nfig /auto fs
⁠D .1.4. /etc /s ys c o nfig /c lo c k

737
737
737
740
742

⁠D .1.5. /etc /s ys c o nfig /d hc p d
⁠D .1.6 . /etc /s ys c o nfig /firs tb o o t
⁠D .1.7. /etc /s ys c o nfig /i18 n
⁠D .1.8 . /etc /s ys c o nfig /init
⁠D .1.9 . /etc /s ys c o nfig /ip 6 tab les -c o nfig

742
742
743
743
745

⁠D .1.10 . /etc /s ys c o nfig /keyb o ard
⁠D .1.11. /etc /s ys c o nfig /ld ap
⁠D .1.12. /etc /s ys c o nfig /named
⁠D .1.13. /etc /s ys c o nfig /netwo rk
⁠D .1.14. /etc /s ys c o nfig /ntp d

746
747
748
748
749

⁠D .1.15. /etc /s ys c o nfig /q uag g a
⁠D .1.16 . /etc /s ys c o nfig /rad vd
⁠D .1.17. /etc /s ys c o nfig /s amb a
⁠D .1.18 . /etc /s ys c o nfig /s as lauthd
⁠D .1.19 . /etc /s ys c o nfig /s elinux

750
751
751
752
752

⁠D .1.20 . /etc /s ys c o nfig /s end mail
⁠D .1.21. /etc /s ys c o nfig /s p amas s as s in
⁠D .1.22. /etc /s ys c o nfig /s q uid
⁠D .1.23. /etc /s ys c o nfig /s ys tem-c o nfig -us ers

753
753
753
754

13

Deployment G uide
⁠D .1.23. /etc /s ys c o nfig /s ys tem-c o nfig -us ers
⁠D .1.24. /etc /s ys c o nfig /vnc s ervers

754
754

⁠ .1.25. /etc /s ys c o nfig /xinetd
D
⁠D .2. Direc to ries in the /etc /s ys c o nfig / Direc to ry
⁠D .3. Ad d itio nal Res o urc es
⁠D .3.1. Ins talled Do c umentatio n

754
755
756
756

. .ppendix
⁠A
. . . . . . . E.
..T
. .he
. . proc
. . . . .File
. . . Syst
. . . . em
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7. 57
...........
⁠E .1. A Virtual File Sys tem
757
⁠E .1.1. Viewing Virtual Files
⁠E .1.2. Chang ing Virtual Files
⁠E .2. To p -level Files within the p ro c File Sys tem
⁠E .2.1. /p ro c /b ud d yinfo
⁠E .2.2. /p ro c /c md line

14

757
758
758
759
759

⁠E .2.3. /p ro c /c p uinfo
⁠E .2.4. /p ro c /c ryp to
⁠E .2.5. /p ro c /d evic es
⁠E .2.6 . /p ro c /d ma
⁠E .2.7. /p ro c /exec d o mains

76 0
76 0
76 1
76 2
76 2

⁠E .2.8 . /p ro c /fb
⁠E .2.9 . /p ro c /files ys tems
⁠E .2.10 . /p ro c /interrup ts
⁠E .2.11. /p ro c /io mem
⁠E .2.12. /p ro c /io p o rts

76 2
76 2
76 3
76 4
76 4

⁠E .2.13. /p ro c /kc o re
⁠E .2.14. /p ro c /kms g
⁠E .2.15. /p ro c /lo ad avg
⁠E .2.16 . /p ro c /lo c ks
⁠E .2.17. /p ro c /md s tat

76 5
76 5
76 5
76 5
76 6

⁠E .2.18 . /p ro c /meminfo
⁠E .2.19 . /p ro c /mis c
⁠E .2.20 . /p ro c /mo d ules
⁠E .2.21. /p ro c /mo unts
⁠E .2.22. /p ro c /mtrr

76 6
76 8
76 8
76 9
76 9

⁠E .2.23. /p ro c /p artitio ns
⁠E .2.24. /p ro c /s lab info
⁠E .2.25. /p ro c /s tat
⁠E .2.26 . /p ro c /s wap s
⁠E .2.27. /p ro c /s ys rq -trig g er

770
770
771
772
772

⁠E .2.28 . /p ro c /up time
⁠E .2.29 . /p ro c /vers io n
⁠E .3. Direc to ries within /p ro c /
⁠E .3.1. Pro c es s Direc to ries
⁠E .3.2. /p ro c /b us /

773
773
773
773
775

⁠E .3.3. /p ro c /b us /p c i
⁠E .3.4. /p ro c /d river/
⁠E .3.5. /p ro c /fs
⁠E .3.6 . /p ro c /irq /
⁠E .3.7. /p ro c /net/

776
777
777
777
777

⁠E .3.8 . /p ro c /s c s i/
⁠E .3.9 . /p ro c /s ys /
⁠E .3.10 . /p ro c /s ys vip c /
⁠E .3.11. /p ro c /tty/
⁠E .3.12. /p ro c /PID/

778
78 0
78 9
78 9
79 0

T able of Cont ent s
⁠E .4. Us ing the s ys c tl Co mmand
⁠E .5. Ad d itio nal Res o urc es
⁠Ins tallab le Do c umentatio n

79 1
79 2
79 2

. .ppendix
⁠A
. . . . . . . F.
. . Revision
. . . . . . . . Hist
. . . . ory
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7. 9. 3. . . . . . . . . .
⁠I.ndex
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7.9. 4. . . . . . . . . .

15

Deployment G uide

⁠Part I. Basic System Configuration
This part covers basic system administration tasks such as keyboard configuration, date and time
configuration, managing users and groups, and gaining privileges.

16

⁠Chapt er 1 . Keyboard Configurat ion

Chapter 1. Keyboard Configuration
This chapter describes how to change the keyboard layout, as well as how to add the K eyb o ard
In d icat o r applet to the panel. It also covers the option to enforce a typing break, and explains both
advantages and disadvantages of doing so.

1.1. Changing t he Keyboard Layout
The installation program has allowed you to configure a keyboard layout for your system. However,
the default settings may not always suit your current needs. To configure a different keyboard layout
after the installation, use the K eyb o ard Pref eren ces tool.
To open K eyb o ard Layo u t Pref eren ces, select Syst em → Pref eren ces → K eyb o ard from the
panel, and click the Layo u t s tab.

Fig u re 1.1. K eyb o ard Layo u t Pref eren ces
You will be presented with a list of available layouts. To add a new one, click the Ad d button below
the list, and you will be prompted to choose which layout you want to add.

17

Deployment G uide

Fig u re 1.2. C h o o sin g a layo u t
Currently, there are two ways how to choose the keyboard layout: you can either find it by the country
it is associated with (the B y co u n t ry tab), or you can select it by language (the B y lan g u ag e tab).
In either case, first select the desired country or language from the C o u n t ry or Lan g u ag e pulldown
menu, then specify the variant from the Varian t s menu. The preview of the layout changes
immediately. To confirm the selection, click Ad d .

Fig u re 1.3. Select in g t h e d ef au lt layo u t
The layout should appear in the list. To make it the default, select the radio button next to its name.
The changes take effect immediately. Note that there is a text-entry field at the bottom of the window
where you can safely test your settings. Once you are satisfied, click C lo se to close the window.

18

⁠Chapt er 1 . Keyboard Configurat ion

Fig u re 1.4 . T est in g t h e layo u t

Disable separate layout for each window
By default, changing the keyboard layout affects the active window only. This means that if
you change the layout and switch to another window, this window will use the old one, which
might be confusing. To turn this behavior off, clear the Sep arat e layo u t f o r each win d o w
checkbox.

D oing this has its drawbacks though, as you will no longer be able to choose the default
layout by selecting the radio button as shown in Figure 1.3, “ Selecting the default layout” . To
make the layout the default, drag it to the beginning of the list.

1.2. Adding t he Keyboard Layout Indicat or
If you want to see what keyboard layout you are currently using, or you would like to switch between
different layouts with a single mouse click, add the K eyb o ard In d icat o r applet to the panel. To do
so, right-click the empty space on the main panel, and select the Ad d t o Pan el option from the
pulldown menu.

19

Deployment G uide

Fig u re 1.5. Ad d in g a n ew ap p let
You will be presented with a list of available applets. Scroll through the list (or start typing
“ keyboard” into the search field at the top of the window), select K eyb o ard In d icat o r, and click the
Ad d button.

Fig u re 1.6 . Select in g t h e K eyb o ard In d icat o r
The applet appears immediately, displaying the shortened name of the country the current layout is
associated with. To display the actual variant, hover the pointer over the applet icon.

20

⁠Chapt er 1 . Keyboard Configurat ion

Fig u re 1.7. T h e K eyb o ard In d icat o r ap p let

1.3. Set t ing Up a T yping Break
Typing for a long period of time can be not only tiring, but it can also increase the risk of serious
health problems, such as carpal tunnel syndrome. One way of preventing this is to configure the
system to enforce typing breaks. To do so, select Syst em → Pref eren ces → K eyb o ard from the
panel, click the T yp in g B reak tab, and select the Lo ck screen t o en f o rce t yp in g b reak
checkbox.

Fig u re 1.8. T yp in g B reak Pro p ert ies

21

Deployment G uide

To increase or decrease the allowed typing time before the break is enforced, click the up or down
button next to the Wo rk in t erval last s label respectively. You can do the same with the B reak
in t erval last s setting to alter the length of the break itself. Finally, select the Allo w p o st p o n in g o f
b reaks checkbox if you want to be able to delay the break in case you need to finish the work. The
changes take effect immediately.

Fig u re 1.9 . T akin g a b reak
Next time you reach the time limit, you will be presented with a screen advising you to take a break,
and a clock displaying the remaining time. If you have enabled it, the Po st p o n e B reak button will
be located at the bottom right corner of the screen.

22

⁠Chapt er 2 . Dat e and T ime Configurat ion

Chapter 2. Date and Time Configuration
This chapter covers setting the system date and time in Red Hat Enterprise Linux, both manually and
using the Network Time Protocol (NTP), as well as setting the adequate time zone. Two methods are
covered: setting the date and time using the D at e/T ime Pro p ert ies tool, and doing so on the
command line.

2.1. Dat e/T ime Propert ies T ool
The D at e/T ime Pro p ert ies tool allows the user to change the system date and time, to configure the
time zone used by the system, and to set up the Network Time Protocol daemon to synchronize the
system clock with a time server. Note that to use this application, you must be running the X Window
System (see Appendix C, The X Window System for more information on this topic).
To start the tool, select Syst em → Ad min ist rat io n → D at e & T ime from the panel, or type the
syst em- co n f ig - d at e command at a shell prompt (e.g., xterm or GNOME Terminal). Unless you are
already authenticated, you will be prompted to enter the superuser password.

Fig u re 2.1. Au t h en t icat io n Q u ery

2.1.1. Dat e and T ime Propert ies
As shown in Figure 2.2, “ D ate and Time Properties” , the D at e/T ime Pro p ert ies tool is divided into
two separate tabs. The tab containing the configuration of the current date and time is shown by
default.

23

Deployment G uide

Fig u re 2.2. D at e an d T ime Pro p ert ies
To set up your system manually, follow these steps:
1. Change the current date. Use the arrows to the left and right of the month and year to change
the month and year respectively. Then click inside the calendar to select the day of the month.
2. Change the current time. Use the up and down arrow buttons beside the H o u r, Min u t e, and
Seco n d , or replace the values directly.
Click the O K button to apply the changes and exit the application.

2.1.2. Net work T ime Prot ocol Propert ies
If you prefer an automatic setup, select the checkbox labeled Syn ch ro n iz e d at e an d t ime o ver
t h e n et wo rk instead. This will display the list of available NTP servers as shown in Figure 2.3,
“ Network Time Protocol Properties” .

24

⁠Chapt er 2 . Dat e and T ime Configurat ion

Fig u re 2.3. N et wo rk T ime Pro t o co l Pro p ert ies
Here you can choose one of the predefined servers, edit a predefined server by clicking the Ed it
button, or add a new server name by clicking Ad d . In the Ad van ced O p t io n s, you can also select
whether you want to speed up the initial synchronization of the system clock, or if you wish to use a
local time source.

Note
Your system does not start synchronizing with the NTP server until you click the O K button at
the bottom of the window to confirm your changes.
Click the O K button to apply any changes made to the date and time settings and exit the
application.

2.1.3. T ime Zone Propert ies
To configure the system time zone, click the T ime Z o n e tab as shown in Figure 2.4, “ Time Z one
Properties” .

25

Deployment G uide

Fig u re 2.4 . T ime Z o n e Pro p ert ies
There are two common approaches to the time zone selection:
1. Using the interactive map. Click “ zoom in” and “ zoom out” buttons next to the map, or click on
the map itself to zoom into the selected region. Then choose the city specific to your time
zone. A red X appears and the time zone selection changes in the list below the map.
2. Use the list below the map. To make the selection easier, cities and countries are grouped
within their specific continents. Note that non-geographic time zones have also been added
to address needs in the scientific community.
If your system clock is set to use UTC, select the Syst em clo ck u ses U T C option. UTC stands for
the Universal Time, Coordinated, also known as Greenwich Mean Time (GMT). Other time zones are
determined by adding or subtracting from the UTC time.
Click O K to apply the changes and exit the program.

2.2. Command Line Configurat ion
In case your system does not have the D at e/T ime Pro p ert ies tool installed, or the X Window Server
is not running, you will have to change the system date and time on the command line. Note that in
order to perform actions described in this section, you have to be logged in as a superuser:

26

⁠Chapt er 2 . Dat e and T ime Configurat ion

~]$ su Password:

2.2.1. Dat e and T ime Set up
The d at e command allows the superuser to set the system date and time manually:
1. Change the current date. Type the command in the following form at a shell prompt, replacing
the YYYY with a four-digit year, MM with a two-digit month, and DD with a two-digit day of the
month:
~]# d at e + % D - s YYYY-MM-DD
For example, to set the date to 2 June 2010, type:
~]# d at e + % D - s 2010- 06 - 02
2. Change the current time. Use the following command, where HH stands for an hour, MM is a
minute, and SS is a second, all typed in a two-digit form:
~]# d at e + % T - s HH:MM:SS
If your system clock is set to use UTC (Coordinated Universal Time), add the following option:
~]# d at e + % T - s HH:MM:SS - u
For instance, to set the system clock to 11:26 PM using the UTC, type:
~]# d at e + % T - s 23:26 :00 - u
You can check your current settings by typing d at e without any additional argument:

Examp le 2.1. D isp layin g t h e cu rren t d at e an d t ime
~]$ d at e
Wed Jun 2 11:58:48 CEST 2010

2.2.2. Net work T ime Prot ocol Set up
As opposed to the manual setup described above, you can also synchronize the system clock with a
remote server over the Network Time Protocol (NTP). For the one-time synchronization only, use the
n t p d at e command:
1. Firstly, check whether the selected NTP server is accessible:
~]# n t p d at e - q server_address
For example:
~]# n t p d at e - q 0.rh el.p o o l.n t p .o rg

27

Deployment G uide

2. When you find a satisfactory server, run the n t p d at e command followed by one or more
server addresses:
~]# n t p d at e server_address...
For instance:
~]# n t p d at e 0.rh el.p o o l.n t p .o rg 1.rh el.p o o l.n t p .o rg
Unless an error message is displayed, the system time should now be set. You can check the
current by setting typing d at e without any additional arguments as shown in Section 2.2.1,
“ D ate and Time Setup” .
3. In most cases, these steps are sufficient. Only if you really need one or more system services
to always use the correct time, enable running the n t p d at e at boot time:
~]# ch kco n f ig n t p d at e o n
For more information about system services and their setup, see Chapter 11, Services and
Daemons.

Note
If the synchronization with the time server at boot time keeps failing, i.e., you find a
relevant error message in the /var/lo g /b o o t .lo g system log, try to add the following
line to /et c/sysco n f ig /n et wo rk:
NETWORKWAIT=1

However, the more convenient way is to set the n t p d daemon to synchronize the time at boot time
automatically:
1. Open the NTP configuration file /et c/n t p .co n f in a text editor such as vi or n an o , or create
a new one if it does not already exist:
~]# n an o /et c/n t p .co n f
2. Now add or edit the list of public NTP servers. If you are using Red Hat Enterprise Linux 6, the
file should already contain the following lines, but feel free to change or expand these
according to your needs:
server
server
server
server

0.rhel.pool.ntp.org iburst
1.rhel.pool.ntp.org iburst
2.rhel.pool.ntp.org iburst
3.rhel.pool.ntp.org iburst

The ib u rst directive at the end of each line is to speed up the initial synchronization. As of
Red Hat Enterprise Linux 6.5 it is added by default. If upgrading from a previous minor
release, and your /et c/n t p .co n f file has been modified, then the upgrade to Red Hat
Enterprise Linux 6.5 will create a new file /et c/n t p .co n f .rp mn ew and will not alter the

28

⁠Chapt er 2 . Dat e and T ime Configurat ion

existing /et c/n t p .co n f file.
3. Once you have the list of servers complete, in the same file, set the proper permissions, giving
the unrestricted access to localhost only:
restrict default kod nomodify notrap nopeer noquery
restrict -6 default kod nomodify notrap nopeer noquery
restrict 127.0.0.1
restrict -6 ::1
4. Save all changes, exit the editor, and restart the NTP daemon:
~]# service n t p d rest art
5. Make sure that n t p d is started at boot time:
~]# ch kco n f ig n t p d o n

29

Deployment G uide

Chapter 3. Managing Users and Groups
3.1. What Users and Groups Are
The control of users and groups is a core element of Red Hat Enterprise Linux system administration.
The user of the system is either a human being or an account used by specific applications identified
by a unique numerical identification number called user ID (UID ). Users within a group can have read
permissions, write permissions, execute permissions or any combination of read/write/execute
permissions for files owned by that group.
Red Hat Enterprise Linux supports access control lists (ACLs) for files and directories which allow
permissions for specific users outside of the owner to be set. For more information about this feature,
see the Access Control Lists chapter of the Red Hat Enterprise Linux 6 Storage Administration Guide.
A group is an organization unit tying users together for a common purpose, which can be reading
permissions, writing permission, or executing permission for files owned by that group. Similar to
UID , each group is associated with a group ID (GID ).

Note
Red Hat Enterprise Linux reserves user and group ID s below 500 for system users and
groups. By default, the U ser Man ag er does not display the system users. Reserved user and
group ID s are documented in the setup rpm package. To view the documentation, use this
command:
cat /u sr/sh are/d o c/set u p - 2.8.14 /u id g id
The recommended practice is to assign non-reserved ID s starting at 5,000, as the reserved
range can increase in the future.
A user who creates a file is also the owner and primary group owner of that file.

Note
Each member of the system is a member of at least one group, a primary group. A
supplementary group is an additional group for accessing files owned by this group. A user
can temporarily change what group is his effective primary group with the n ewg rp command,
allowing all files created afterwords (until the effect is undone) to have a specific group
ownership
The file is assigned separate read, write, and execute permissions for the owner, the group, and
everyone else. The file owner can be changed only by ro o t , and access permissions can be
changed by both the ro o t user and file owner.
The setting which determines what permissions are applied to a newly created file or directory is
called a u mask and is configured in the /et c/b ash rc file. Note that only the user who created the file
or directory is allowed to make modifications.
Individual users on the system are provided with sh ad o w p asswo rd s. Encrypted password hashes
are stored in the /et c/sh ad o w file, which is readable only by the ro o t user, and store information
about password aging as well. Policies for specific account are stored in /et c/sh ad o w.

30

⁠Chapt er 3. Managing Users and G roups

/et c/lo g in .d ef s and /et c/d ef au lt /u serad d define default values, which are stored in /et c/sh ad o w
when the account is being created. For more information on shadowing, see the pwconv(8) and
grpconv(8) manual pages.

Note
Most utilities provided by the shadow-utils package work properly whether or not shadow
passwords are enabled, apart from the commands and utilities which create or modify
password aging information. These are: ch ag e and g p asswd utilities, the u sermo d
command with the - e or - f options and the u serad d command with the - e or - f options. For
more information on how to enable password aging, see the Red Hat Enterprise Linux 6 Security
Guide.

3.2. Managing Users via t he User Manager Applicat ion
The U ser Man ag er application allows you to view, modify, add, and delete local users and groups
in the graphical user interface.
T o st art t h e U ser Man ag er ap p licat io n :
From the toolbar, select Syst em → Ad min ist rat io n → U sers an d G ro u p s.
Or, type syst em- co n f ig - u sers at the shell prompt.

Note
Unless you have superuser privileges, the application will prompt you to authenticate as ro o t .

3.2.1. Viewing Users
In order to display the main window of the U ser Man ag er to view users, from the toolbar of U ser
Man ag er select Ed it → Pref eren ces. If you wish to view all the users, that is including system
users, clear the H id e syst em u sers an d g ro u p s checkbox.
The U sers tab provides a list of local users along with additional information about their user ID ,
primary group, home directory, login shell, and full name.

31

Deployment G uide

Fig u re 3.1. Viewin g U sers
To find a specific user, type the first few letters of the name in the Search f ilt er field and either press
En t er, or click the Ap p ly f ilt er button. You can also sort the items according to any of the available
columns by clicking the column header.

3.2.2. Adding a New User
If there is a new user you need to add to the system, follow this procedure:
1. Click the Ad d U ser button.
2. Enter the user name and full name in the appropriate fields
3. Type the user's password in the Passwo rd and C o n f irm Passwo rd fields. The password
must be at least six characters long.

Note
For safety reasons, choose a long password not based on a dictionary term; use a
combination of letters, numbers, and special characters.
4. Select a login shell for the user from the Lo g in Sh ell dropdown list or accept the default
value of /b in /b ash .
5. Clear the C reat e h o me d irect o ry checkbox if you choose not to create the home directory
for a new user in /h o me/username/.

32

⁠Chapt er 3. Managing Users and G roups

You can also change this home directory by editing the content of the H o me D irect o ry text
box. Note that when the home directory is created, default configuration files are copied into it
from the /et c/skel/ directory.
6. Clear the C reat e a p rivat e g ro u p f o r t h e u ser checkbox if you do not want a unique
group with the same name as the user to be created. User private group (UPG) is a group
assigned to a user account to which that user exclusively belongs, which is used for
managing file permissions for individual users.
7. Specify a user ID for the user by selecting Sp ecif y u ser ID man u ally. If the option is not
selected, the next available user ID above 500 is assigned to the new user.
8. Click the O K button to complete the process.
Look at the sample Ad d N ew U ser dialog box configuration:

To configure more advanced user properties, such as password expiration, modify the user's
properties after adding the user.

3.2.3. Modifying User Propert ies
1. Select the user from the user list by clicking once on the user name.

33

Deployment G uide

2. Click Pro p ert ies from the toolbar or choose File → Pro p ert ies from the dropdown menu.

Fig u re 3.2. U ser Pro p ert ies
3. There are four tabs you can update to your preferences. When you have finished, click the
O K button to save your changes.

3.3. Managing Groups via t he User Manager Applicat ion
3.3.1. Viewing Groups
In order to display the main window of U ser Man ag er to view groups, from the toolbar select Ed it →
Pref eren ces. If you wish to view all the groups, clear the H id e syst em u sers an d g ro u p s
checkbox.
The G ro u p s tab provides a list of local groups with information about their group ID and group
members as you can see in the picture below.

34

⁠Chapt er 3. Managing Users and G roups

Fig u re 3.3. Viewin g G ro u p s
To find a specific group, type the first few letters of the name in the Search f ilt er field and either
press En t er, or click the Ap p ly f ilt er button. You can also sort the items according to any of the
available columns by clicking the column header.

3.3.2. Adding a New Group
If there is a new group you need to add to the system, follow this procedure:
1. Select Ad d G ro u p from the User Manager toolbar:

Fig u re 3.4 . N ew G ro u p
2. Type the name of the new group.
3. Specify the group ID (GID ) for the new group by checking the Sp ecif y g ro u p ID man u ally
checkbox.

35

Deployment G uide

4. Select the GID . Note that Red Hat Enterprise Linux also reserves group ID s lower than 500 for
system groups.
5. Click O K to create the group. The new group appears in the group list.

3.3.3. Modifying Group Propert ies
1. Select the group from the group list by clicking on its name.
2. Click Pro p ert ies from the toolbar or choose File → Pro p ert ies from the dropdown menu.

Fig u re 3.5. G ro u p Pro p ert ies
3. The G ro u p U sers tab displays the list of group members. Use this tab to add or remove
users from the group. Click O K to save your changes.

3.4 . Managing Users via Command-Line T ools
When managing users via command line, the following commands are used: u serad d , u sermo d ,
u serd el, or p asswd . The files affected include /et c/p asswd which stores user accounts
information and /et c/sh ad o w, which stores secure user account information.

3.4 .1. Creat ing Users
The u serad d utility creates new users and adds them to the system. Following the short procedure
below, you will create a default user account with its UID , automatically create a home directory
where default user settings will be stored, /h o me/username/, and set the default shell to /b in /b ash .

36

⁠Chapt er 3. Managing Users and G roups

1. Run the following command at a shell prompt as ro o t substituting username with the name of
your choice:
u serad d username
2. By setting a password unlock the account to make it accessible. Type the password twice
when the program prompts you to.
p asswd

Examp le 3.1. C reat in g a U ser wit h D ef au lt Set t in g s
~]# useradd robert
~]# passwd robert
Changing password for user robert
New password:
Re-type new password:
passwd: all authentication tokens updated successfully.
Running the u serad d ro b ert command creates an account named ro b ert . If you run cat
/et c/p asswd to view the content of the /et c/p asswd file, you can learn more about the new user
from the line displayed to you:
robert:x:502:502::/home/robert:/bin/bash
ro b ert has been assigned a UID of 502, which reflects the rule that the default UID values from 0
to 499 are typically reserved for system accounts. GID , group ID of U ser Privat e G ro u p , equals
to UID . The home directory is set to /h o me/ro b ert and login shell to /b in /b ash . The letter x
signals that shadow passwords are used and that the hashed password is stored in
/et c/sh ad o w.
If you want to change the basic default setup for the user while creating the account, you can choose
from a list of command-line options modifying the behavior of u serad d (see the u serad d (8) man
page for the whole list of options). As you can see from the basic syntax of the command, you can
add one or more options:
u serad d [ option(s)] username
As a system administrator, you can use the - c option to specify, for example, the full name of the user
when creating them. Use - c followed by a string, which adds a comment to the user:
u serad d - c "string" username

Examp le 3.2. Sp ecif yin g a U ser' s Fu ll N ame wh en C reat in g a U ser
~]# useradd -c " Robert Smith" robert
~]# cat /etc/passwd
robert:x:502:502:Robert Smith:/home/robert:/bin/bash

37

Deployment G uide

A user account has been created with user name ro b ert , sometimes called the login name, and
full name Robert Smith.
If you do not want to create the default /h o me/username/ directory for the user account, set a different
one instead of it. Execute the command below:
u serad d - d home_directory

Examp le 3.3. Ad d in g a U ser wit h n o n - d ef au lt H o me D irect o ry
~]# useradd -d /home/dir_1 robert
ro b ert 's home directory is now not the default /h o me/ro b ert but /h o me/d ir_1/.
If you do not want to create the home directory for the user at all, you can do so by running u serad d
with the - M option. However, when such a user logs into a system that has just booted and their
home directory does not exist, their login directory will be the root directory. If such a user logs into a
system using the su command, their login directory will be the current directory of the previous user.
u serad d - M username
If you need to copy a directory content to the /h o me directory while creating a new user, make use of
the - m and - k options together followed by the path.

Examp le 3.4 . C reat in g a U ser wh ile C o p yin g C o n t en t s t o t h e H o me D irect o ry
The following command copies the contents of a directory named /d ir_1 to /h o me/jan e, which is
the default home directory of a new user jan e:
~]# useradd -m -k /dir_1 jane

As a system administrator, you may need to create a temporary account. Using the u serad d
command, this means creating an account for a certain amount of time only and disabling it at a
certain date. This is a particularly useful setting as there is no security risk resulting from forgetting to
delete a certain account. For this, the - e option is used with the specified expire_date in the YYYY-MMD D format.

Note
D o not confuse account expiration and password expiration. Account expiration is a
particular date, after which it is impossible to log in to the account in any way, as the account
no longer exists. Password expiration, the maximum password age and date of password
creation or last password change, is the date, when it is not possible to log in using the
password (but other ways exist, such as loging in by SSH key).

u serad d - e YYYY-MM-DD username

38

⁠Chapt er 3. Managing Users and G roups

Examp le 3.5. Set t in g t h e Acco u n t Exp irat io n D at e
~]# useradd -e 2015-11-05 emily
The account emily will be created now and automatically disabled on 5 November, 2015.
User's login shell defaults to /b in /b ash , but can be changed by the - s option to any other shell
different from bash, ksh, csh, tsh, for example.
u serad d - s login_shell username

Examp le 3.6 . Ad d in g a U ser wit h N o n - d ef au lt Sh ell
~]# useradd -s /bin/ksh robert
This command creates the user ro b ert which has the /b in /ksh shell.
The - r option creates a system account, which is an account for administrative use that has some,
but not all, root privileges. Such accounts have a UID lower than the value of UID_MIN defined in
/et c/lo g in .d ef s, typically 500 and above for ordinary users.
u serad d - r username

3.4 .2. At t aching New Users t o Groups
The u serad d command creates a U ser Privat e G ro u p (UPG, a group assigned to a user account
to which that user exclusively belongs) whenever a new user is added to the system and names the
group after the user. For example, when the account ro b ert is created, an UPG named ro b ert is
created at the same time, the only member of which is the user ro b ert .
If you do not want to create a U ser Privat e G ro u p for a user for whatever reason, execute the
u serad d command with the following option:
u serad d - N username
Instead of automatically creating UPG or not creating it at all, you can specify the user's group
membership with - g and - G options. While the - g option specifies the primary group membership, - G
refers to supplementary groups into which the user is also included. The group names you specify
must already exist on the system.

Examp le 3.7. Ad d in g a U ser t o a G ro u p
~]# useradd -g " friends" -G " family,schoolmates" emily
The u serad d - g "f rien d s" - G "f amily,sch o o lmat es" emily command creates the user emily
but emily's primary group is set to f rien d s as specified by the - g option. emily is also a group
member of the supplementary groups f amily and sch o o lmat es.
Provided the user already exists and you want to add them to certain supplementary group(s), use
the u sermo d command with the - G option and a list of groups divided by commas, no spaces:

39

Deployment G uide

u sermo d - G group_1,group_2,group_3

3.4 .3. Updat ing Users' Aut hent icat ion
When running the basic u serad d username command, the password is automatically set to never
expire (see the /et c/sh ad o w file).
If you wish to change this, use p asswd , the standard utility for administering the /et c/p asswd file.
The syntax of the p asswd command look as follows:
p asswd option(s) username
You can, for example, lock the specified account. The locking is performed by rendering the
encrypted password into an invalid string by prefixing the encrypted string with an the exclamation
mark (!). If you later find a reason to unlock the account, p asswd has a reverse operation for
locking. Only ro o t can carry out these two operations.
p asswd - l username
p asswd - u username

Examp le 3.8. U n lo ckin g a U ser Passwo rd
~]# passwd -l robert
Locking password for user robert.
passwd: Success
~]# passwd -u robert
passwd: Warning: unlocked password would be empty
passwd: Unsafe operation (use -f to force)
At first, the - l option locks ro b ert 's account password successfully. However, running the
p asswd - u command does not unlock the password because by default p asswd refuses to
create a passwordless account.
If you wish a password for an account to expire, run p asswd with the - e option. The user will be
forced to change the password during the next login attempt:
p asswd - e username
As far as the password lifetime is concerned, setting the minimum time between password changes is
useful for forcing the user to really change the password. The system administrator can set the
minimum (the - n option) and the maximum (the - x option) lifetimes. To inform the user about their
password expiration, use the - w option. All these options must be accompanied with the number of
days and can be run as ro o t only.

Examp le 3.9 . Ad ju st in g Ag in g D at a f o r U ser Passwo rd s
~]# passwd -n 10 -x 60 -w 3 jane

40

⁠Chapt er 3. Managing Users and G roups

The above command has set the minimum password lifetime to 10 days, the maximum password
lifetime to 60, and the number of days jan e will begin receiving warnings in advance that her
password will expire to 3 day.
Later, when you cannot remember the password setting, make use of the - S option which outputs a
short information for you to know the status of the password for a given account:
~]# passwd -S jane
jane LK 2014-07-22 10 60 3 -1 (Password locked.)

You can also set the number of days after a password expires with the u serad d command, which
disables the account permanently. A value of 0 disables the account as soon as the password has
expired, and a value of - 1 disables the feature, that is, the user will have to change his password
when the password expires. The - f option is used to specify the number of days after a password
expires until the account is disabled (but may be unblocked by system administrator):
u serad d - f number-of-days username
For more information on the p asswd command see the p asswd (1) man page.

3.4 .4 . Modifying User Set t ings
When a user already exists and you need to specify any of the options now, use the u sermo d
command. The logic of using u sermo d is identical to u serad d as well as its syntax:
u sermo d option(s) username
If you need to change the user's user name, use the - l option with the new user name (or login).

Examp le 3.10. C h an g in g U ser' s Lo g in
~]# usermod -l " emily-smith" emily
The - l option changes the name of the user from the login emily to the new login, emily- smit h .
Nothing else is changed. In particular, emily's home directory name (/h o me/emily) remains the
same unless it is changed manually to reflect the new user name.
In the similar way you can change the user's UID or user's home directory. StStudy the example
below:

Note
Find all files owned by the specified UID in system and change their owner. D o the same for
Access Control List (ACL) referring to the UID . It is recommended to check there are no running
processes as they keep running with the old UID .

Examp le 3.11. C h an g in g U ser' s U ID an d H o me D irect o ry

41

Deployment G uide

~]# usermod -a -u 699 -d /home/dir_2 robert
The command with - a - u and - d options changes the settings of user ro b ert . Now, his ID is 699
instead of 501, and his home directory is no longer /h o me/ro b ert but /h o me/d ir_2.
With the u sermo d command you can also move the content of the user's home directory to a new
location, or lock the account by locking its password.

Examp le 3.12. C h an g in g U ser' s
~]# usermod -m -d /home/jane -L jane
In this sample command, the - m and - d options used together move the content of jan e's home
directory to the /h o me/d ir_3 directory. The - L option locks the access to jan e's account by
locking its password.
For the whole list of options to be used with the u sermo d command, see the u sermo d (8) man page
or run u sermo d - - h elp on the command line.

3.4 .5. Delet ing Users
If you want to remove a user account from the system, use the u serd el command on the command
line as ro o t .
u serd el username
Combining u serd el with the - r option removes files in the user's home directory along with the home
directory itself and the user's mail spool. Files located in other file systems have to be searched for
and deleted manually.
u serd el - r username

Note
The - r option is relatively safer, and thus recommended, compared to - f which forces the
removal of the user account even if the user is still logged in.

3.4 .6. Displaying Comprehensive User Informat ion
When administering users and groups on your system, you need a good tool to monitor their
configuration and activity on the system. Red Hat Enterprise Linux 6 provides you with the lslo g in s
command-line utility, which gives you a comprehensive overview of users and groups, not only
regarding user or group account configuration but also their activity on the system.
The general syntax of lslo g in s is the following:
lslo g in s [ OPTIONS]

42

⁠Chapt er 3. Managing Users and G roups

where OPTIONS can be one or more available options and their related parameters. See the
lslo g in s(1) manual page or the output of the lslo g in s - - h elp command for the complete list of
available options and their usage.
The lslo g in s utility gives versatile information in a variety of formats based on the chosen options.
The following examples introduce the most basic as well as some of the most useful combinations.
Running the lslo g in s command without any options shows default information about all system
and user accounts on the system. Specifically, their UID , user name, and GECOS information, as well
as information about the user's last login to the system, and whether their password is locked or
login by password disabled.

Examp le 3.13. D isp layin g b asic in f o rmat io n ab o u t all acco u n t s o n t h e syst em
~]# lslo g in s
UID USER
PWD -LOCK PWD -D ENY LAST-LOGIN GECOS
0 root
0
0
root
1 bin
0
1
bin
2 daemon
0
1
daemon
3 adm
0
1
adm
4 lp
0
1
lp
5 sync
0
1
sync
6 shutdown
0
1 Jul21/16:20 shutdown
7 halt
0
1
halt
8 mail
0
1
mail
10 uucp
0
1
uucp
11 operator
0
1
operator
12 games
0
1
games
13 gopher
0
1
gopher
14 ftp
0
1
FTP User
29 rpcuser
0
1
RPC Service User
32 rpc
0
1
Rpcbind D aemon
38 ntp
0
1
42 gdm
0
1
48 apache
0
1
Apache
68 haldaemon
0
1
HAL daemon
69 vcsa
0
1
virtual console memory owner
72 tcpdump
0
1
74 sshd
0
1
Privilege-separated SSH
81 dbus
0
1
System message bus
89 postfix
0
1
99 nobody
0
1
Nobody
113 usbmuxd
0
1
usbmuxd user
170 avahi-autoipd
0
1
Avahi IPv4LL Stack
173 abrt
0
1
497 pulse
0
1
PulseAudio System D aemon
498 saslauth
0
1
Saslauthd user
499 rtkit
0
1
RealtimeKit
500 jsmith
0
0 10:56:12 John Smith
501 jdoe
0
0 12:13:53 John D oe
502 esmith
0
0 12:59:05 Emily Smith
503 jeyre
0
0 12:22:14 Jane Eyre
65534 nfsnobody
0
1
Anonymous NFS User

43

Deployment G uide

To display detailed information about a single user, run the lslo g in s LOGIN command, where
LOGIN is either a UID or a user name. The following example displays detailed information about
Jo h n D o e's account and his activity on the system:

Examp le 3.14 . D isp layin g d et ailed in f o rmat io n ab o u t a sin g le acco u n t
~]# lslo g in s jd o e
Username:
jdoe
UID :
501
Gecos field:
John D oe
Home directory:
/home/jdoe
Shell:
/bin/bash
No login:
no
Password is locked:
no
Password no required:
no
Login by password disabled:
no
Primary group:
jdoe
GID :
501
Supplementary groups:
users
Supplementary group ID s:
100
Last login:
12:13:53
Last terminal:
pts/3
Last hostname:
192.168.100.1
Hushed:
no
Password expiration warn interval: 7
Password changed:
Aug01/02:00
Maximal change time:
99999
Password expiration:
Sep01/02:00
Selinux context:
unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023

If you use the - - lo g in s= LOGIN option, you can display information about a group of accounts that
are specified as a list of UID s or user names. Specifying the - - o u t p u t = COLUMNS option, where
COLUMNS is a list of available output parameters, you can customize the output of the lslo g in s
command. For example, the following command shows login activity of the users root, jsmith, jdoe,
and esmith:

Examp le 3.15. D isp layin g sp ecif ic in f o rmat io n ab o u t a g ro u p o f u sers
~]# lslo g in s - - lo g in s= 0,500,jd o e,esmit h \
> - - o u t p u t = U ID ,U SER ,LAST - LO G IN ,LAST - T T Y,FAILED - LO G IN ,FAILED - T T Y
UID USER LAST-LOGIN LAST-TTY FAILED -LOGIN FAILED -TTY
0 root
500 jsmith 10:56:12 pts/2
501 jdoe 12:13:53 pts/3
502 esmith 15:46:16 pts/3 15:46:09 ssh:notty

The lslo g in s utility also distinguishes between system and user accounts. To address system
accounts in your query, use the - - syst em- accs option. To address user accounts, use the - - u seraccs. For example, the following command displays information about supplementary groups and
password expirations for all user accounts:

44

⁠Chapt er 3. Managing Users and G roups

Examp le 3.16 . D isp layin g in f o rmat io n ab o u t su p p lemen t ary g ro u p s an d p asswo rd
exp irat io n f o r all u ser acco u n t s
~]# lslo g in s - - u ser- accs - - su p p - g ro u p s - - acc- exp irat io n
UID USER
GID GROUP SUPP-GID S SUPP-GROUPS PWD -WARN PWD -MIN PWD MAX PWD -CHANGE
PWD -EXPIR
0 root
0 root
7
99999 Jul21/02:00
500 jsmith
500 jsmith 1000,100 staff,users
7
99999 Jul21/02:00
501 jdoe
501 jdoe
100
users
7
99999 Aug01/02:00
Sep01/02:00
502 esmith
502 esmith 100
users
7
99999 Aug01/02:00
503 jeyre
503 jeyre 1000,100 staff,users
7
99999 Jul28/02:00
Sep01/02:00
65534 nfsnobody 65534 nfsnobody
Jul21/02:00

The ability to format the output of lslo g in s commands according to the user's needs makes
lslo g in s an ideal tool to use in scripts and for automatic processing. For example, the following
command returns a single string that represents the time and date of the last login. This string can be
passed as input to another utility for further processing.

Examp le 3.17. D isp layin g a sin g le p iece o f in f o rmat io n wit h o u t t h e h ead in g
~]# lslo g in s - - lo g in s= jsmit h - - o u t p u t = LAST - LO G IN - - t ime- f o rmat = iso | t ail - 1
2014-08-06T10:56:12+0200

3.5. Managing Groups via Command-Line T ools
Groups are a useful tool for permitting co-operation between different users. There is a set of
commands for operating with groups such as g ro u p ad d , g ro u p mo d , g ro u p d el, or g p asswd .
The files affected include /et c/g ro u p which stores group account information and /et c/g sh ad o w,
which stores secure group account information.

3.5.1. Creat ing Groups
To add a new group to the system with default settings, the g ro u p ad d command is run at the shell
prompt as ro o t .
g ro u p ad d group_name

Examp le 3.18. C reat in g a G ro u p wit h D ef au lt Set t in g s
~]# groupadd friends
The g ro u p ad d command creates a new group called f rien d s. You can read more information
about the group from the newly-created line in the /et c/g ro u p file:
classmates:x:30005:

45

Deployment G uide

Automatically, the group f rien d s is attached with a unique GID (group ID ) of 30005 and is not
attached with any users. Optionally, you can set a password for a group by running g p asswd
groupname.
Alternatively, you can add command options with specific settings.
g ro u p ad d option(s) groupname
If you, for example, want to specify the numerical value of the group's ID (GID ) when creating the
group, run the g ro u p ad d command with the - g option. Remember that this value must be unique
(unless the - o option is used) and the value must be non-negative.
g ro u p ad d - g GID

Examp le 3.19 . C reat in g a G ro u p wit h Sp ecif ied G ID
The command below creates a group named sch o o lmat es and sets GID of 60002 for it:
~]# groupadd -g 60002 schoolmates

When used with - g and GID already exists, g ro u p ad d refuses to create another group with existing
GID . As a workaround, use the - f option, with which g ro u p ad d creates a group, but with a different
GID.
g ro u p ad d - f GID
You may also create a system group by attaching the - r option to the g ro u p ad d command. System
groups are used for system purposes, which practically means that GID is allocated from 1 to 499
within the reserved range of 999.
g ro u p ad d - r group_name
For more information on g ro u p ad d , see the groupadd(8) man pages.

3.5.2. At t aching Users t o Groups
If you want to add an existing user to the named group, you can make use of the g p asswd
command.
g p asswd - a username which_group_to_edit
To remove a user from the named group, run:
g p asswd - d username which_group_to_edit
To set the list of group members, write the user names after the - - memb ers option dividing them with
commas and no spaces:
g p asswd - - memb ers username_1,username_2 which_group_to_edit

46

⁠Chapt er 3. Managing Users and G roups

3.5.3. Updat ing Group Aut hent icat ion
The g p asswd command administers /et c/g ro u p and /et c/g sh ad o w files. Note that this command
works only if run by a group administrator.
Who is a group administrator? A group administrator can add and delete users as well as set,
change, or remove the group password. A group can have more than one group administrator. The
ro o t user can add group administrators with the g p asswd - A users groupname where users is a
comma-separated list of existing users you want to be group administrators (without any spaces
between commas).
For changing a group's password, run the g p asswd command with the relevant group name. You
will be prompted to type the new password of the group.
g p asswd groupname

Examp le 3.20. C h an g in g a G ro u p Passwo rd
~]# gpasswd crowd
Changing password for group crowd
New password:
Re-enter new password:
The password for the group cro wd has been changed.
You can also remove the password from the named group by using the - r option.
g p asswd - r sch o o lmat es

3.5.4 . Modifying Group Set t ings
When a group already exists and you need to specify any of the options now, use the g ro u p mo d
command. The logic of using g ro u p mo d is identical to g ro u p ad d as well as its syntax:
g ro u p mo d option(s) groupname
To change the group ID of a given group, use the g ro u p mo d command in the following way:
g ro u p mo d - g GID_NEW which_group_to_edit

Note
Find all files owned by the specified GID in system and change their owner. D o the same for
Access Control List (ACL) referring to the GID . It is recommended to check there are no running
processes as they keep running with the old GID .
To change the name of the group, run the following on the command line. The name of the group will
be changed from GROUP_NAME to NEW_GROUP_NAME name.

47

Deployment G uide

g ro u p mo d - n new_groupname groupname

Examp le 3.21. C h an g in g a G ro u p ' s N ame
The following command changes the name of the group sch o o lmat es to cro wd :
~]# groupmod -n crowd schoolmates

3.5.5. Delet ing Groups
The g ro u p d el command modifies the system account files, deleting all entries that see the group.
The named group must exist when you execute this command.
g ro u p d el groupname

3.6. Addit ional Resources
See the following resources for more information about managing users and groups.

3.6.1. Inst alled Document at ion
For information about various utilities for managing users and groups, see the following manual
pages:
ch ag e(1) — A command to modify password aging policies and account expiration.
g p asswd (1) — A command to administer the /et c/g ro u p file.
g ro u p ad d (8) — A command to add groups.
g rp ck(8) — A command to verify the /et c/g ro u p file.
g ro u p d el(8) — A command to remove groups.
g ro u p mo d (8) — A command to modify group membership.
p wck(8) — A command to verify the /et c/p asswd and /et c/sh ad o w files.
p wco n v(8) — A tool to convert standard passwords to shadow passwords.
p wu n co n v(8) — A tool to convert shadow passwords to standard passwords.
u serad d (8) — A command to add users.
u serd el(8) — A command to remove users.
u sermo d (8) — A command to modify users.
For information about related configuration files, see:
g ro u p (5) — The file containing group information for the system.
p asswd (5) — The file containing user information for the system.

48

⁠Chapt er 3. Managing Users and G roups

sh ad o w(5) — The file containing passwords and account expiration information for the system.
lo g in .d ef s(5) - The file containing shadow password suite configuration.
u serad d (8) - For /et c/d ef au lt /u serad d , section “ Changing the default values” in manual page.

49

Deployment G uide

Chapter 4. Gaining Privileges
System administrators (and in some cases users) will need to perform certain tasks with
administrative access. Accessing the system as ro o t is potentially dangerous and can lead to
widespread damage to the system and data. This chapter covers ways to gain administrative
privileges using setuid programs such as su and su d o . These programs allow specific users to
perform tasks which would normally be available only to the root user while maintaining a higher
level of control and system security.
See the Red Hat Enterprise Linux 6 Security Guide for more information on administrative controls,
potential dangers and ways to prevent data loss resulting from improper use of privileged access.

4 .1. T he su Command
When a user executes the su command, they are prompted for the root password and, after
authentication, are given a root shell prompt.
Once logged in via the su command, the user is the root user and has absolute administrative
access to the system ⁠ [1] . In addition, once a user has become root, it is possible for them to use the
su command to change to any other user on the system without being prompted for a password.
Because this program is so powerful, administrators within an organization may wish to limit who
has access to the command.
One of the simplest ways to do this is to add users to the special administrative group called wheel.
To do this, type the following command as root:
u sermo d - G wh eel 
In the previous command, replace  with the user name you want to add to the wh eel
group.
You can also use the U ser Man ag er to modify group memberships, as follows. Note: you need
Administrator privileges to perform this procedure.
1. Click the Syst em menu on the Panel, point to Ad min ist rat io n and then click U sers an d
G ro u p s to display the User Manager. Alternatively, type the command syst em- co n f ig u sers at a shell prompt.
2. Click the U sers tab, and select the required user in the list of users.
3. Click Pro p ert ies on the toolbar to display the User Properties dialog box (or choose
Pro p ert ies on the File menu).
4. Click the G ro u p s tab, select the checkbox for the wheel group, and then click O K .
See Section 3.2, “ Managing Users via the User Manager Application” for more information about the
U ser Man ag er.
After you add the desired users to the wh eel group, it is advisable to only allow these specific users
to use the su command. To do this, you will need to edit the PAM configuration file for su :
/et c/p am.d /su . Open this file in a text editor and remove the comment (#) from the following line:
#auth

50

required

pam_wheel.so use_uid

⁠Chapt er 4 . G aining Privileges

This change means that only members of the administrative group wh eel can switch to another user
using the su command.

Note
The ro o t user is part of the wh eel group by default.

4 .2. T he sudo Command
The su d o command offers another approach to giving users administrative access. When trusted
users precede an administrative command with su d o , they are prompted for their own password.
Then, when they have been authenticated and assuming that the command is permitted, the
administrative command is executed as if they were the root user.
The basic format of the su d o command is as follows:
su d o 
In the above example,  would be replaced by a command normally reserved for the root
user, such as mo u n t .
The su d o command allows for a high degree of flexibility. For instance, only users listed in the
/et c/su d o ers configuration file are allowed to use the su d o command and the command is
executed in the user's shell, not a root shell. This means the root shell can be completely disabled as
shown in the Red Hat Enterprise Linux 6 Security Guide.
Each successful authentication using the su d o is logged to the file /var/lo g /messag es and the
command issued along with the issuer's user name is logged to the file /var/lo g /secu re. Should you
require additional logging, use the p am_t t y_au d it module to enable TTY auditing for specified
users by adding the following line to your /et c/p am.d /syst em- au t h file:
session required pam_tty_audit.so disable= enable=
where pattern represents a comma-separated listing of users with an optional use of globs. For
example, the following configuration will enable TTY auditing for the root user and disable it for all
other users:
session required pam_tty_audit.so disable=* enable=root
Another advantage of the su d o command is that an administrator can allow different users access to
specific commands based on their needs.
Administrators wanting to edit the su d o configuration file, /et c/su d o ers, should use the visu d o
command.
To give someone full administrative privileges, type visu d o and add a line similar to the following in
the user privilege specification section:
juan ALL=(ALL) ALL
This example states that the user, ju an , can use su d o from any host and execute any command.
The example below illustrates the granularity possible when configuring su d o :

51

Deployment G uide

% users localhost=/sbin/shutdown -h now
This example states that any user can issue the command /sb in /sh u t d o wn - h n o w as long as it is
issued from the console.
The man page for su d o ers has a detailed listing of options for this file.

Important
There are several potential risks to keep in mind when using the su d o command. You can
avoid them by editing the /et c/su d o ers configuration file using visu d o as described above.
Leaving the /et c/su d o ers file in its default state gives every user in the wh eel group unlimited
root access.
By default, su d o stores the sudoer's password for a five minute timeout period. Any
subsequent uses of the command during this period will not prompt the user for a
password. This could be exploited by an attacker if the user leaves his workstation
unattended and unlocked while still being logged in. This behavior can be changed by
adding the following line to the /et c/su d o ers file:
D efaults

timestamp_timeout=

where  is the desired timeout length in minutes. Setting the  to 0 causes
su d o to require a password every time.
If a sudoer's account is compromised, an attacker can use su d o to open a new shell with
administrative privileges:
su d o /b in /b ash
Opening a new shell as root in this or similar fashion gives the attacker administrative
access for a theoretically unlimited amount of time, bypassing the timeout period specified
in the /et c/su d o ers file and never requiring the attacker to input a password for su d o
again until the newly opened session is closed.

4 .3. Addit ional Resources
While programs allowing users to gain administrative privileges are a potential security risk, security
itself is beyond the scope of this particular book. You should therefore see sources listed below for
more information regarding security and privileged access.

Inst alled Document at ion
su (1) - the manual page for su provides information regarding the options available with this
command.
su d o (8) - the manual page for su d o includes a detailed description of this command as well as
a list of options available for customizing su d o 's behavior.
p am(8) - the manual page describing the use of Pluggable Authentication Modules for Linux.

52

⁠Chapt er 4 . G aining Privileges

Online Document at ion
Red Hat Enterprise Linux 6 Security Guide - The Security Guide provides a more in-depth look at
potential security issues pertaining to setuid programs as well as techniques used to alleviate
these risks.
Red Hat Enterprise Linux 6 Managing Single Sign-On and Smart Cards - This guide provides,
among other things, a detailed description of Pluggable Authentication Modules (PAM), their
configuration and usage.

[1] This ac c es s is s till s ub jec t to the res tric tio ns imp o s ed b y SELinux, if it is enab led .

53

Deployment G uide

⁠Part II. Subscription and Support
To receive updates to the software on a Red Hat Enterprise Linux system it must be subscribed to the
Red Hat Content Delivery Network (CD N) and the appropriate repositories enabled. This part describes
how to subscribe a system to the Red Hat Content D elivery Network.
Red Hat provides support via the Customer Portal, and you can access this support directly from the
command line using the R ed H at Su p p o rt T o o l. This part describes the use of this command-line
tool.

54

⁠Chapt er 5. Regist ering t he Syst em and Managing Subscript ions

Chapter 5. Registering the System and Managing Subscriptions
The subscription service provides a mechanism to handle Red Hat software inventory and allows
you to install additional software or update already installed programs to newer versions using the
yu m or Packag eK it package managers. In Red Hat Enterprise Linux 6 the recommended way to
register your system and attach subscriptions is to use Red Hat Subscription Management.

Note
It is also possible to register the system and attach subscriptions after installation during the
firstboot process. For detailed information about firstboot see the Firstboot chapter in the
Installation Guide for Red Hat Enterprise Linux 6. Note that firstboot is only available on
systems after a graphical installation or after a kickstart installation where a desktop and the X
window system were installed and graphical login was enabled.

5.1. Regist ering t he Syst em and At t aching Subscript ions
Complete the following steps to register your system and attach one or more subscriptions using
Red Hat Subscription Management. Note that all su b scrip t io n - man ag er commands are supposed
to be run as ro o t .
1. Run the following command to register your system. You will be prompted to enter your user
name and password. Note that the user name and password are the same as your login
credentials for Red Hat Customer Portal.
su b scrip t io n - man ag er reg ist er
2. D etermine the pool ID of a subscription that you requires. To do so, type the following at a
shell prompt to display a list of all subscriptions that are available for your system:
su b scrip t io n - man ag er list - - availab le
For each available subscription, this command displays its name, unique identifier,
expiration date, and other details related to your subscription. The pool ID is listed on a line
beginning with Po o l ID .
3. Attach the appropriate subscription to your system by entering a command as follows:
su b scrip t io n - man ag er at t ach - - p o o l= pool_id
Replace pool_id with the pool ID you determined in the previous step. To verify the list of
subscriptions your system has currently attached, at any time, run:
su b scrip t io n - man ag er list - - co n su med

55

Deployment G uide

Note
The au t o - at t ach option automatically attaches compatible subscriptions to the system so
that you can register the system and attach subscriptions in one step by executing the
following command:
su b scrip t io n - man ag er reg ist er - - au t o - at t ach

For more information on how to register your system using Red Hat Subscription Management and
associate it with subscriptions, see the Red Hat Subscription Management collection of guides.

5.2. Managing Soft ware Reposit ories
When a system is subscribed to the Red Hat Content D elivery Network, a repository file is created in
the /et c/yu m.rep o s.d / directory. To verify that, use yu m to list all enabled repositories:
yu m rep o list
Red Hat Subscription Management also allows you to manually enable or disable software
repositories provided by Red Hat. To list all available repositories, use the following command:
su b scrip t io n - man ag er rep o s - - list
The repository names depend on the specific version of Red Hat Enterprise Linux you are using and
are in the following format:
rhel-variant-rhscl-version-rpms
rhel-variant-rhscl-version-debug-rpms
rhel-variant-rhscl-version-source-rpms
Where variant is the Red Hat Enterprise Linux system variant (server or wo rkst at io n ), and version is
the Red Hat Enterprise Linux system version (6 or 7), for example:
rhel-server-rhscl-6 -eus-rpms
rhel-server-rhscl-6 -eus-source-rpms
rhel-server-rhscl-6 -eus-debug-rpms
To enable a repository, enter a command as follows:
su b scrip t io n - man ag er rep o s - - en ab le repository
Replace repository with a name of the repository to enable.
Similarly, to disable a repository, use the following command:
su b scrip t io n - man ag er rep o s - - d isab le repository
Section 7.4, “ Configuring Yum and Yum Repositories” provides detailed information about managing
software repositories using yu m.

56

⁠Chapt er 5. Regist ering t he Syst em and Managing Subscript ions

5.3. Removing Subscript ions
To remove a particular subscription, complete the following steps.
1. D etermine the serial number of the subscription you want to remove by listing information
about already attached subscriptions:
su b scrip t io n - man ag er list - - co n su med
The serial number is the number listed as serial. For instance, 74 4 9 9 3814 251016 831 in
the example below:
SKU:
ES0113909
Contract:
01234567
Account:
1234567
Serial:
744993814251016831
Pool ID :
8a85f9894bba16dc014bccdd905a5e23
Active:
False
Quantity Used: 1
Service Level: SELF-SUPPORT
Service Type:
L1-L3
Status D etails:
Subscription Type: Standard
Starts:
02/27/2015
Ends:
02/27/2016
System Type:
Virtual
2. Enter a command as follows to remove the selected subscription:
su b scrip t io n - man ag er remo ve - - serial= serial_number
Replace serial_number with the serial number you determined in the previous step.
To remove all subscriptions attached to the system, run the following command:
su b scrip t io n - man ag er remo ve - - all

5.4 . Addit ional Resources
For more information on how to register your system using Red Hat Subscription Management and
associate it with subscriptions, see the resources listed below.

Inst alled Document at ion
su b scrip t io n - man ag er(8) — the manual page for Red Hat Subscription Management provides
a complete list of supported options and commands.

Relat ed Books
Red Hat Subscription Management collection of guides — These guides contain detailed
information how to use Red Hat Subscription Management.
Installation Guide — see the Firstboot chapter for detailed information on how to register during

57

Deployment G uide

the firstboot process.

Online Resources
Red Hat Access Labs — The Red Hat Access Labs includes a “ Registration Assistant” .

See Also
Chapter 4, Gaining Privileges documents how to gain administrative privileges by using the su and
su d o commands.
Chapter 7, Yum provides information about using the yu m packages manager to install and
update software.
Chapter 8, PackageKit provides information about using the Packag eK it package manager to
install and update software.

58

⁠Chapt er 6 . Accessing Support Using t he Red Hat Support T ool

Chapter 6. Accessing Support Using the Red Hat Support Tool
The R ed H at Su p p o rt T o o l, in the redhat-support-tool package, can function as both an interactive
shell and as a single-execution program. It can be run over SSH or from any terminal. It enables, for
example, searching the Red Hat Knowledgebase from the command line, copying solutions directly
on the command line, opening and updating support cases, and sending files to Red Hat for
analysis.

6.1. Inst alling t he Red Hat Support T ool
The R ed H at Su p p o rt T o o l is installed by default on Red Hat Enterprise Linux. If required, to
ensure that it is, enter the following command as ro o t :
~]# yu m in st all red h at - su p p o rt - t o o l

6.2. Regist ering t he Red Hat Support T ool Using t he Command Line
To register the Red Hat Support Tool to the customer portal using the command line, proceed as
follows:
1.

~]# red h at - su p p o rt - t o o l co n f ig u ser username
Where username is the user name of the Red Hat Customer Portal account.

2.

~]# red h at - su p p o rt - t o o l co n f ig p asswo rd
Please enter the password for username:

6.3. Using t he Red Hat Support T ool in Int eract ive Shell Mode
To start the tool in interactive mode, enter the following command:
~]$ red h at - su p p o rt - t o o l
Welcome to the Red Hat Support Tool.
Command (? for help):
The tool can be run as an unprivileged user, with a consequently reduced set of commands, or as
ro o t .
The commands can be listed by entering the ? character. The program or menu selection can be
exited by entering the q or e character. You will be prompted for your Red Hat Customer Portal user
name and password when you first search the Knowledgebase or support cases. Alternately, set the
user name and password for your Red Hat Customer Portal account using interactive mode, and
optionally save it to the configuration file.

6.4 . Configuring t he Red Hat Support T ool
When in interactive mode, the configuration options can be listed by entering the command co n f ig - h elp :

59

Deployment G uide

~]# red h at - su p p o rt - t o o l
Welcome to the Red Hat Support Tool.
Command (? for help): co n f ig - - h elp
Usage: config [options] config.option 
Use the 'config' command to set or get configuration file values.
Options:
-h, --help show this help message and exit
-g, --global Save configuration option in /etc/redhat-support-tool.conf.
-u, --unset Unset configuration option.
The configuration file options which can be set are:
user
: The Red Hat Customer Portal user.
password : The Red Hat Customer Portal password.
debug : CRITICAL, ERROR, WARNING, INFO, or D EBUG
url
: The support services URL. D efault=https://api.access.redhat.com
proxy_url : A proxy server URL.
proxy_user: A proxy server user.
proxy_password: A password for the proxy server user.
ssl_ca : Path to certificate authorities to trust during communication.
kern_debug_dir: Path to the directory where kernel debug symbols should be downloaded and
cached. D efault=/var/lib/redhat-support-tool/debugkernels
Examples:
- config user
- config user my-rhn-username
- config --unset user
Pro ced u re 6 .1. R eg ist erin g t h e R ed H at Su p p o rt T o o l U sin g In t eract ive Mo d e
To register the Red Hat Support Tool to the customer portal using interactive mode, proceed as
follows:
1. Start the tool by entering the following command:
~]# red h at - su p p o rt - t o o l
2. Enter your Red Hat Customer Portal user name:
Command (? for help): co n f ig u ser username
To save your user name to the global configuration file, add the - g option.
3. Enter your Red Hat Customer Portal password:
Command (? for help): co n f ig p asswo rd
Please enter the password for username:

6.4 .1. Saving Set t ings t o t he Configurat ion Files
The R ed H at Su p p o rt T o o l, unless otherwise directed, stores values and options locally in the
home directory of the current user, using the ~ /.red h at - su p p o rt - t o o l/red h at - su p p o rt t o o l.co n f configuration file. If required, it is recommended to save passwords to this file because it

60

⁠Chapt er 6 . Accessing Support Using t he Red Hat Support T ool

is only readable by that particular user. When the tool starts, it will read values from the global
configuration file /et c/red h at - su p p o rt - t o o l.co n f and from the local configuration file. Locally
stored values and options take precedence over globally stored settings.

Warning
It is recommended n o t to save passwords in the global /et c/red h at - su p p o rt - t o o l.co n f
configuration file because the password is just b ase6 4 encoded and can easily be decoded.
In addition, the file is world readable.
To save a value or option to the global configuration file, add the - g , - - g lo b al option as follows:
Command (? for help): co n f ig setting - g value

Note
In order to be able to save settings globally, using the - g , - - g lo b al option, the R ed H at
Su p p o rt T o o l must be run as ro o t because normal users do not have the permissions
required to write to /et c/red h at - su p p o rt - t o o l.co n f .
To remove a value or option from the local configuration file, add the - u , - - u n set option as follows:
Command (? for help): co n f ig setting - u value
This will clear, unset, the parameter from the tool and fall back to the equivalent setting in the global
configuration file, if available.

Note
When running as an unprivileged user, values stored in the global configuration file cannot
be removed using the - u , - - u n set option, but they can be cleared, unset, from the current
running instance of the tool by using the - g , - - g lo b al option simultaneously with the - u , - u n set option. If running as ro o t , values and options can be removed from the global
configuration file using - g , - - g lo b al simultaneously with the - u , - - u n set option.

6.5. Opening and Updat ing Support Cases Using Int eract ive Mode
Pro ced u re 6 .2. O p en in g a N ew Su p p o rt C ase U sin g In t eract ive Mo d e
To open a new support case using interactive mode, proceed as follows:
1. Start the tool by entering the following command:
~]# red h at - su p p o rt - t o o l
2. Enter the o p en case command:

61

Deployment G uide

Command (? for help): o p en case
3. Follow the on screen prompts to select a product and then a version.
4. Enter a summary of the case.
5. Enter a description of the case and press C t rl+D on an empty line when complete.
6. Select a severity of the case.
7. Optionally chose to see if there is a solution to this problem before opening a support case.
8. Confirm you would still like to open the support case.
Support case 0123456789 has successfully been opened
9. Optionally chose to attach an SOS report.
10. Optionally chose to attach a file.
Pro ced u re 6 .3. Viewin g an d U p d at in g an Exist in g Su p p o rt C ase U sin g In t eract ive Mo d e
To view and update an existing support case using interactive mode, proceed as follows:
1. Start the tool by entering the following command:
~]# red h at - su p p o rt - t o o l
2. Enter the g et case command:
Command (? for help): g et case case-number
Where case-number is the number of the case you want to view and update.
3. Follow the on screen prompts to view the case, modify or add comments, and get or add
attachments.
Pro ced u re 6 .4 . Mo d if yin g an Exist in g Su p p o rt C ase U sin g In t eract ive Mo d e
To modify the attributes of an existing support case using interactive mode, proceed as follows:
1. Start the tool by entering the following command:
~]# red h at - su p p o rt - t o o l
2. Enter the mo d if ycase command:
Command (? for help): mo d if ycase case-number
Where case-number is the number of the case you want to view and update.
3. The modify selection list appears:
Type the number of the attribute to modify or 'e' to return to the previous menu.
1 Modify Type
2 Modify Severity

62

⁠Chapt er 6 . Accessing Support Using t he Red Hat Support T ool

3 Modify Status
4 Modify Alternative-ID
5 Modify Product
6 Modify Version
End of options.
Follow the on screen prompts to modify one or more of the options.
4. For example, to modify the status, enter 3:
Selection: 3
1 Waiting on Customer
2 Waiting on Red Hat
3 Closed
Please select a status (or 'q' to exit):

6.6. Viewing Support Cases on t he Command Line
Viewing the contents of a case on the command line provides a quick and easy way to apply
solutions from the command line.
To view an existing support case on the command line, enter a command as follows:
~]# red h at - su p p o rt - t o o l g et case case-number
Where case-number is the number of the case you want to download.

6.7. Addit ional Resources
The Red Hat Knowledgebase article Red Hat Support Tool has additional information, examples, and
video tutorials.

63

Deployment G uide

⁠Part III. Installing and Managing Software
All software on a Red Hat Enterprise Linux system is divided into RPM packages, which can be
installed, upgraded, or removed. This part focuses on product subscriptions and entitlements, and
describes how to manage packages on Red Hat Enterprise Linux using both Yu m and the
Packag eK it suite of graphical package management tools.

64

⁠Chapt er 7 . Yum

Chapter 7. Yum
Yu m is the Red Hat package manager that is able to query for information about available
packages, fetch packages from repositories, install and uninstall them, and update an entire system
to the latest available version. Yum performs automatic dependency resolution on packages you are
updating, installing, or removing, and thus is able to automatically determine, fetch, and install all
available dependent packages.
Yum can be configured with new, additional repositories, or package sources, and also provides
many plug-ins which enhance and extend its capabilities. Yum is able to perform many of the same
tasks that R PM can; additionally, many of the command-line options are similar. Yum enables easy
and simple package management on a single machine or on groups of them.

Secure package management with GPG-signed packages
Yum provides secure package management by enabling GPG (Gnu Privacy Guard; also
known as GnuPG) signature verification on GPG-signed packages to be turned on for all
package repositories (i.e. package sources), or for individual repositories. When signature
verification is enabled, Yum will refuse to install any packages not GPG-signed with the
correct key for that repository. This means that you can trust that the R PM packages you
download and install on your system are from a trusted source, such as Red Hat, and were not
modified during transfer. See Section 7.4, “ Configuring Yum and Yum Repositories” for details
on enabling signature-checking with Yum, or Section B.3, “ Checking a Package's Signature”
for information on working with and verifying GPG-signed R PM packages in general.
Yum also enables you to easily set up your own repositories of R PM packages for download and
installation on other machines.
Learning Yum is a worthwhile investment because it is often the fastest way to perform system
administration tasks, and it provides capabilities beyond those provided by the Packag eK it
graphical package management tools. See Chapter 8, PackageKit for details on using Packag eK it .

Yum and superuser privileges
You must have superuser privileges in order to use yu m to install, update or remove packages
on your system. All examples in this chapter assume that you have already obtained
superuser privileges by using either the su or su d o command.

7.1. Checking For and Updat ing Packages
7.1.1. Checking For Updat es
To see which installed packages on your system have updates available, use the following
command:
yu m ch eck- u p d at e
For example:

65

Deployment G uide

~]# yu m ch eck- u p d at e
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
PackageKit.x86_64
0.5.8-2.el6
rhel
PackageKit-glib.x86_64
0.5.8-2.el6
rhel
PackageKit-yum.x86_64
0.5.8-2.el6
rhel
PackageKit-yum-plugin.x86_64
0.5.8-2.el6
rhel
glibc.x86_64
2.11.90-20.el6
rhel
glibc-common.x86_64
2.10.90-22
rhel
kernel.x86_64
2.6.31-14.el6
rhel
kernel-firmware.noarch
2.6.31-14.el6
rhel
rpm.x86_64
4.7.1-5.el6
rhel
rpm-libs.x86_64
4.7.1-5.el6
rhel
rpm-python.x86_64
4.7.1-5.el6
rhel
udev.x86_64
147-2.15.el6
rhel
yum.noarch
3.2.24-4.el6
rhel
The packages in the above output are listed as having updates available. The first package in the
list is Packag eK it , the graphical package manager. The line in the example output tells us:
Packag eK it — the name of the package
x86 _6 4 — the CPU architecture the package was built for
0.5.8 — the version of the updated package to be installed
rh el — the repository in which the updated package is located
The output also shows us that we can update the kernel (the kernel package), Yum and RPM
themselves (the yum and rpm packages), as well as their dependencies (such as the kernel-firmware,
rpm-libs, and rpm-python packages), all using yu m.

7.1.2. Updat ing Packages
You can choose to update a single package, multiple packages, or all packages at once. If any
dependencies of the package (or packages) you update have updates available themselves, then
they are updated too.

Updat ing a Single Package
To update a single package, run the following command as ro o t :
yu m u p d at e package_name
For example, to update the udev package, type:
~]# yu m u p d at e u d ev
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Setting up Update Process
Resolving D ependencies
--> Running transaction check
---> Package udev.x86_64 0:147-2.15.el6 set to be updated

66

⁠Chapt er 7 . Yum

--> Finished D ependency Resolution
D ependencies Resolved
===========================================================================
Package
Arch
Version
Repository Size
===========================================================================
Updating:
udev
x86_64
147-2.15.el6
rhel
337 k
Transaction Summary
===========================================================================
Install
0 Package(s)
Upgrade
1 Package(s)
Total download size: 337 k
Is this ok [y/N]:
This output contains several items of interest:
1. Lo ad ed p lu g in s: p ro d u ct - id , ref resh - p ackag ekit , su b scrip t io n - man ag er — yu m
always informs you which Yum plug-ins are installed and enabled. See Section 7.5, “ Yum
Plug-ins” for general information on Yum plug-ins, or to Section 7.5.3, “ Plug-in D escriptions”
for descriptions of specific plug-ins.
2. u d ev.x86 _6 4 — you can download and install new udev package.
3. yu m presents the update information and then prompts you as to whether you want it to
perform the update; yu m runs interactively by default. If you already know which transactions
the yu m command plans to perform, you can use the - y option to automatically answer yes
to any questions that yu m asks (in which case it runs non-interactively). However, you
should always examine which changes yu m plans to make to the system so that you can
easily troubleshoot any problems that might arise.
If a transaction does go awry, you can view Yum's transaction history by using the yu m
h ist o ry command as described in Section 7.3, “ Working with Transaction History” .

Updating and installing kernels with Yum
yu m always installs a new kernel in the same sense that R PM installs a new kernel when you
use the command rp m - i kern el. Therefore, you do not need to worry about the distinction
between installing and upgrading a kernel package when you use yu m: it will do the right thing,
regardless of whether you are using the yu m u p d at e or yu m in st all command.
When using R PM, on the other hand, it is important to use the rp m - i kern el command (which
installs a new kernel) instead of rp m - u kern el (which replaces the current kernel). See
Section B.2.2, “ Installing and Upgrading” for more information on installing/upgrading
kernels with R PM.

Updat ing All Package s and T he ir De pe nde ncie s
To update all packages and their dependencies, enter yu m u p d at e (without any arguments):
yu m u p d at e

67

Deployment G uide

Updat ing Se curit y-Re lat e d Package s
D iscovering which packages have security updates available and then updating those packages
quickly and easily is important. Yum provides the plug-in for this purpose. The secu rit y plug-in
extends the yu m command with a set of highly-useful security-centric commands, subcommands and
options. See Section 7.5.3, “ Plug-in D escriptions” for specific information.

Updat ing Package s Aut o m at ically
It is also possible to set up periodical automatic updates for your packages. For this purpose,
Red Hat Enterprise Linux 6 uses the yum-cron package. It provides a Yum interface for the cro n
daemon and downloads metadata from your package repositories. With the yu m- cro n service
enabled, the user can schedule an automated daily Yum update as a cron job.

Note
The yum-cron package is provided by the Optional subscription channel. See Section 7.4.8,
“ Adding the Optional and Supplementary Repositories” for more information on Red Hat
additional channels.
To install yu m- cro n issue the following command:
~]# yu m in st all yu m- cro n
By default, the yu m- cro n service is disabled and needs to be activated and started manually:
~]# ch kco n f ig yu m- cro n o n
~]# service yu m- cro n st art
To verify the status of the service, run the following command:
~]# service yu m- cro n st at u s
The script included in the yum-cron package can be configured to change the extent and frequency
of the updates, as well as to send notifications to e-mail. To customize yu m- cro n , edit the
/et c/sysco n f ig /yu m- cro n file.
Additional details and instructions for yu m- cro n can be found in the comments within
/et c/sysco n f ig /yu m- cro n and at the yu m- cro n (8) manual page.

7.1.3. Preserving Configurat ion File Changes
You will inevitably make changes to the configuration files installed by packages as you use your
Red Hat Enterprise Linux system. R PM, which Yum uses to perform changes to the system, provides a
mechanism for ensuring their integrity. See Section B.2.2, “ Installing and Upgrading” for details on
how to manage changes to configuration files across package upgrades.

7.1.4 . Upgrading t he Syst em Off-line wit h ISO and Yum

68

⁠Chapt er 7 . Yum

For systems that are disconnected from the Internet or Red Hat Network, using the yu m u p d at e
command with the Red Hat Enterprise Linux installation ISO image is an easy and quick way to
upgrade systems to the latest minor version. The following steps illustrate the upgrading process:
1. Create a target directory to mount your ISO image. This directory is not automatically created
when mounting, so create it before proceeding to the next step. As ro o t , type:
mkd ir mount_dir
Replace mount_dir with a path to the mount directory. Typically, users create it as a
subdirectory in the /med ia/ directory.
2. Mount the Red Hat Enterprise Linux 6 installation ISO image to the previously created target
directory. As ro o t , type:
mo u n t - o lo o p iso_name mount_dir
Replace iso_name with a path to your ISO image and mount_dir with a path to the target
directory. Here, the - o lo o p option is required to mount the file as a block device.
3. Copy the med ia.rep o file from the mount directory to the /et c/yu m.rep o s.d / directory. Note
that configuration files in this directory must have the .repo extension to function properly.
cp mount_dir/med ia.rep o /et c/yu m.rep o s.d /new.repo
This creates a configuration file for the yum repository. Replace new.repo with the filename, for
example rhel6.repo.
4. Edit the new configuration file so that it points to the Red Hat Enterprise Linux installation
ISO. Add the following line into the /et c/yu m.rep o s.d /new.repo file:
baseurl=file://mount_dir
Replace mount_dir with a path to the mount point.
5. Update all yum repositories including /et c/yu m.rep o s.d /new.repo created in previous steps.
As ro o t , type:
yu m u p d at e
This upgrades your system to the version provided by the mounted ISO image.
6. After successful upgrade, you can unmount the ISO image. As ro o t , type:
u mo u n t mount_dir
where mount_dir is a path to your mount directory. Also, you can remove the mount directory
created in the first step. As ro o t , type:
rmd ir mount_dir
7. If you will not use the previously created configuration file for another installation or update,
you can remove it. As ro o t , type:

69

Deployment G uide

rm /et c/yu m.rep o s.d /new.repo

Examp le 7.1. U p g rad in g f ro m R ed H at En t erp rise Lin u x 6 .3 t o 6 .4
Imagine you need to upgrade your system without access to the Internet. To do so, you want to
use an ISO image with the newer version of the system, called for instance R H EL6 .4 - Server20130130.0- x86 _6 4 - D VD 1.iso . A target directory created for mounting is /med ia/rh el6 /. As
ro o t , change into the directory with your ISO image and type:
~]# mo u n t - o lo o p R H EL6 .4 - Server- 20130130.0- x86 _6 4 - D VD 1.iso /med ia/rh el6 /
Then set up a yum repository for your image by copying the med ia.rep o file from the mount
directory:
~]# cp /med ia/rh el6 /med ia.rep o /et c/yu m.rep o s.d /rh el6 .rep o
To make yum recognize the mount point as a repository, add the following line into the
/et c/yu m.rep o s.d /rh el6 .rep o copied in the previous step:
baseurl=file:///media/rhel6/
Now, updating the yum repository will upgrade your system to a version provided by R H EL6 .4 Server- 20130130.0- x86 _6 4 - D VD 1.iso . As ro o t , execute:
~]# yu m u p d at e
When your system is successfully upgraded, you can unmount the image, remove the target
directory and the configuration file:
~]# u mo u n t /med ia/rh el6 /
~]# rmd ir /med ia/rh el6 /
~]# rm /et c/yu m.rep o s.d /rh el6 .rep o

7.2. Packages and Package Groups
7.2.1. Searching Packages
You can search all RPM package names, descriptions and summaries by using the following
command:
yu m search term…
This command displays the list of matches for each term. For example, to list all packages that match
“ meld” or “ kompare” , type:
~]# yu m search meld ko mp are

70

⁠Chapt er 7 . Yum

Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
============================ Matched: kompare =============================
kdesdk.x86_64 : The KD E Software D evelopment Kit (SD K)
Warning: No matches found for: meld
The yu m search command is useful for searching for packages you do not know the name of, but
for which you know a related term.

7.2.2. List ing Packages
yu m list and related commands provide information about packages, package groups, and
repositories.
All of Yum's list commands allow you to filter the results by appending one or more glob expressions
as arguments. Glob expressions are normal strings of characters which contain one or more of the
wildcard characters * (which expands to match any character multiple times) and ? (which expands
to match any one character).

Filtering results with glob expressions
Be careful to escape the glob expressions when passing them as arguments to a yu m
command, otherwise the Bash shell will interpret these expressions as pathname expansions,
and potentially pass all files in the current directory that match the globs to yu m. To make sure
the glob expressions are passed to yu m as intended, either:
escape the wildcard characters by preceding them with a backslash character
double-quote or single-quote the entire glob expression.
See Example 7.2, “ Listing all ABRT addons and plug-ins using glob expressions” and
Example 7.4, “ Listing available packages using a single glob expression with escaped
wildcard characters” for an example usage of both these methods.

yu m list glob_expression…
Lists information on installed and available packages matching all glob expressions.

Examp le 7.2. List in g all AB R T ad d o n s an d p lu g - in s u sin g g lo b exp ressio n s
Packages with various ABRT addons and plug-ins either begin with “ abrt-addon-” , or
“ abrt-plugin-” . To list these packages, type the following at a shell prompt:
~]# yu m list ab rt - ad d o n \* ab rt - p lu g in \*
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Installed Packages
abrt-addon-ccpp.x86_64
1.0.7-5.el6
@rhel
abrt-addon-kerneloops.x86_64
1.0.7-5.el6
@rhel
abrt-addon-python.x86_64
1.0.7-5.el6
@rhel

71

Deployment G uide

abrt-plugin-bugzilla.x86_64
abrt-plugin-logger.x86_64
abrt-plugin-sosreport.x86_64
abrt-plugin-ticketuploader.x86_64

1.0.7-5.el6
1.0.7-5.el6
1.0.7-5.el6
1.0.7-5.el6

@rhel
@rhel
@rhel
@rhel

yu m list all
Lists all installed and available packages.
yu m list in st alled
Lists all packages installed on your system. The rightmost column in the output lists the
repository from which the package was retrieved.

Examp le 7.3. List in g in st alled p ackag es u sin g a d o u b le- q u o t ed g lo b
exp ressio n
To list all installed packages that begin with “ krb” followed by exactly one character and
a hyphen, type:
~]# yu m list in st alled "krb ?- *"
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Installed Packages
krb5-libs.x86_64
1.8.1-3.el6
@rhel
krb5-workstation.x86_64
1.8.1-3.el6
@rhel
yu m list availab le
Lists all available packages in all enabled repositories.

Examp le 7.4 . List in g availab le p ackag es u sin g a sin g le g lo b exp ressio n wit h
escap ed wild card ch aract ers
To list all available packages with names that contain “ gstreamer” and then “ plugin” ,
run the following command:
~]# yu m list availab le g st reamer\*p lu g in \*
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Available Packages
gstreamer-plugins-bad-free.i686
0.10.17-4.el6
rhel
gstreamer-plugins-base.i686
0.10.26-1.el6
rhel
gstreamer-plugins-base-devel.i686
0.10.26-1.el6
rhel
gstreamer-plugins-base-devel.x86_64
0.10.26-1.el6
rhel
gstreamer-plugins-good.i686
0.10.18-1.el6
rhel
yu m g ro u p list
Lists all package groups.
yu m rep o list

72

⁠Chapt er 7 . Yum

Lists the repository ID , name, and number of packages it provides for each enabled
repository.

7.2.3. Displaying Package Informat ion
To display information about one or more packages (glob expressions are valid here as well), use
the following command:
yu m in f o package_name…
For example, to display information about the abrt package, type:
~]# yu m in f o ab rt
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Installed Packages
Name
: abrt
Arch
: x86_64
Version : 1.0.7
Release : 5.el6
Size
: 578 k
Repo
: installed
From repo : rhel
Summary : Automatic bug detection and reporting tool
URL
: https://fedorahosted.org/abrt/
License : GPLv2+
D escription: abrt is a tool to help users to detect defects in applications
: and to create a bug report with all informations needed by
: maintainer to fix it. It uses plugin system to extend its
: functionality.
The yu m in f o package_name command is similar to the rp m - q - - in f o package_name command,
but provides as additional information the ID of the Yum repository the RPM package is found in
(look for the Fro m rep o : line in the output).
You can also query the Yum database for alternative and useful information about a package by
using the following command:
yu md b in f o package_name
This command provides additional information about a package, including the checksum of the
package (and algorithm used to produce it, such as SHA-256), the command given on the command
line that was invoked to install the package (if any), and the reason that the package is installed on
the system (where u ser indicates it was installed by the user, and d ep means it was brought in as a
dependency). For example, to display additional information about the yum package, type:
~]# yu md b in f o yu m
Loaded plugins: product-id, refresh-packagekit, subscription-manager
yum-3.2.27-4.el6.noarch
checksum_data =
23d337ed51a9757bbfbdceb82c4eaca9808ff1009b51e9626d540f44fe95f771
checksum_type = sha256
from_repo = rhel
from_repo_revision = 1298613159

73

Deployment G uide

from_repo_timestamp = 1298614288
installed_by = 4294967295
reason = user
releasever = 6.1
For more information on the yu md b command, see the yu md b (8) manual page.

List ing File s Co nt aine d in a Package
repoquery is a program for querying information from yum repositories similarly to rpm queries. You
can query both package groups and individual packages. To list all files contained in a specific
package, type:
rep o q u ery - - list package_name
Replace package_name with a name of the package you want to inspect. For more information on the
rep o q u ery command, see the rep o q u ery manual page.
To find out which package provides a specific file, you can use the yu m p ro vid es command,
described in Finding which package owns a file

7.2.4 . Inst alling Packages
Yum allows you to install both a single package and multiple packages, as well as a package group
of your choice.

Inst alling Individual Package s
To install a single package and all of its non-installed dependencies, enter a command in the
following form:
yu m in st all package_name
You can also install multiple packages simultaneously by appending their names as arguments:
yu m in st all package_name package_name…
If you are installing packages on a multilib system, such as an AMD 64 or Intel64 machine, you can
specify the architecture of the package (as long as it is available in an enabled repository) by
appending .arch to the package name. For example, to install the sqlite package for i6 86 , type:
~]# yu m in st all sq lit e.i6 86
You can use glob expressions to quickly install multiple similarly-named packages:
~]# yu m in st all p erl- C ryp t - \*
In addition to package names and glob expressions, you can also provide file names to yu m
in st all. If you know the name of the binary you want to install, but not its package name, you can
give yu m in st all the path name:
~]# yu m in st all /u sr/sb in /n amed

74

⁠Chapt er 7 . Yum

yu m then searches through its package lists, finds the package which provides /u sr/sb in /n amed , if
any, and prompts you as to whether you want to install it.

Finding which package owns a file
If you know you want to install the package that contains the n amed binary, but you do not
know in which b in or sb in directory is the file installed, use the yu m p ro vid es command with
a glob expression:
~]# yu m p ro vid es "*b in /n amed "
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
32:bind-9.7.0-4.P1.el6.x86_64 : The Berkeley Internet Name D omain (BIND )
: D NS (D omain Name System) server
Repo
: rhel
Matched from:
Filename : /usr/sbin/named
yu m p ro vid es "*/file_name" is a common and useful trick to find the package(s) that contain
file_name.

Inst alling a Package Gro up
A package group is similar to a package: it is not useful by itself, but installing one pulls a group of
dependent packages that serve a common purpose. A package group has a name and a groupid. The
yu m g ro u p list - v command lists the names of all package groups, and, next to each of them, their
groupid in parentheses. The groupid is always the term in the last pair of parentheses, such as kd ed eskt o p in the following example:
~]# yu m - v g ro u p list kd e\*
Loading " product-id" plugin
Loading " refresh-packagekit" plugin
Loading " subscription-manager" plugin
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Config time: 0.123
Yum Version: 3.2.29
Setting up Group Process
Looking for repo options for [rhel]
rpmdb time: 0.001
group time: 1.291
Available Groups:
KD E D esktop (kde-desktop)
D one
You can install a package group by passing its full group name (without the groupid part) to
g ro u p in st all:
yu m g ro u p in st all group_name
You can also install by groupid:

75

Deployment G uide

yu m g ro u p in st all groupid
You can even pass the groupid (or quoted name) to the in st all command if you prepend it with an
@-symbol (which tells yu m that you want to perform a g ro u p in st all):
yu m in st all @group
For example, the following are alternative but equivalent ways of installing the K D E D eskt o p group:
~]# yu m g ro u p in st all "K D E D eskt o p "
~]# yu m g ro u p in st all kd e- d eskt o p
~]# yu m in st all @ kd e- d eskt o p

7.2.5. Removing Packages
Similarly to package installation, Yum allows you to uninstall (remove in R PM and Yum terminology)
both individual packages and a package group.

Re m o ving Individual Package s
To uninstall a particular package, as well as any packages that depend on it, run the following
command as ro o t :
yu m remo ve package_name…
As when you install multiple packages, you can remove several at once by adding more package
names to the command. For example, to remove totem, rhythmbox, and sound-juicer, type the following
at a shell prompt:
~]# yu m remo ve t o t em rh yt h mb o x so u n d - ju icer
Similar to in st all, remo ve can take these arguments:
package names
glob expressions
file lists
package provides

Removing a package when other packages depend on it
Yum is not able to remove a package without also removing packages which depend on it.
This type of operation can only be performed by R PM, is not advised, and can potentially
leave your system in a non-functioning state or cause applications to misbehave and/or
crash. For further information, see Section B.2.4, “ Uninstalling” in the R PM chapter.

Re m o ving a Package Gro up
You can remove a package group using syntax congruent with the in st all syntax:

76

⁠Chapt er 7 . Yum

yu m g ro u p remo ve group
yu m remo ve @group
The following are alternative but equivalent ways of removing the K D E D eskt o p group:
~]# yu m g ro u p remo ve "K D E D eskt o p "
~]# yu m g ro u p remo ve kd e- d eskt o p
~]# yu m remo ve @ kd e- d eskt o p

Intelligent package group removal
When you tell yu m to remove a package group, it will remove every package in that group,
even if those packages are members of other package groups or dependencies of other
installed packages. However, you can instruct yu m to remove only those packages which are
not required by any other packages or groups by adding the g ro u p remo ve_leaf _o n ly= 1
directive to the [ main ] section of the /et c/yu m.co n f configuration file. For more information
on this directive, see Section 7.4.1, “ Setting [main] Options” .

7.3. Working wit h T ransact ion Hist ory
The yu m h ist o ry command allows users to review information about a timeline of Yum transactions,
the dates and times they occurred, the number of packages affected, whether transactions succeeded
or were aborted, and if the RPM database was changed between transactions. Additionally, this
command can be used to undo or redo certain transactions.

7.3.1. List ing T ransact ions
To display a list of twenty most recent transactions, as ro o t , either run yu m h ist o ry with no
additional arguments, or type the following at a shell prompt:
yu m h ist o ry list
To display all transactions, add the all keyword:
yu m h ist o ry list all
To display only transactions in a given range, use the command in the following form:
yu m h ist o ry list start_id..end_id
You can also list only transactions regarding a particular package or packages. To do so, use the
command with a package name or a glob expression:
yu m h ist o ry list glob_expression…
For example, the list of the first five transactions looks as follows:

77

Deployment G uide

~]# yu m h ist o ry list 1..5
Loaded plugins: product-id, refresh-packagekit, subscription-manager
ID | Login user
| D ate and time | Action(s)
| Altered
------------------------------------------------------------------------------5 | Jaromir ...  | 2011-07-29 15:33 | Install
| 1
4 | Jaromir ...  | 2011-07-21 15:10 | Install
| 1
3 | Jaromir ...  | 2011-07-16 15:27 | I, U
| 73
2 | System 
| 2011-07-16 15:19 | Update
| 1
1 | System 
| 2011-07-16 14:38 | Install
| 1106
history list
All forms of the yu m h ist o ry list command produce tabular output with each row consisting of the
following columns:
ID — an integer value that identifies a particular transaction.
Lo g in u ser — the name of the user whose login session was used to initiate a transaction. This
information is typically presented in the Full Name  form. For transactions that were
not issued by a user (such as an automatic system update), Syst em  is used instead.
D at e an d t ime — the date and time when a transaction was issued.
Act io n ( s) — a list of actions that were performed during a transaction as described in Table 7.1,
“ Possible values of the Action(s) field” .
Alt ered — the number of packages that were affected by a transaction, possibly followed by
additional information as described in Table 7.2, “ Possible values of the Altered field” .
T ab le 7.1. Po ssib le valu es o f t h e Act io n ( s) f ield
Act io n

Ab b reviat i
on

D escrip t io n

D o wn g rad e

D

Erase
In st all
O b so let in g
R ein st all
U p d at e

E
I
O
R
U

At least one package has been downgraded to an older
version.
At least one package has been removed.
At least one new package has been installed.
At least one package has been marked as obsolete.
At least one package has been reinstalled.
At least one package has been updated to a newer version.

T ab le 7.2. Po ssib le valu es o f t h e Alt ered f ield
Symb o l

D escrip t io n

<

Before the transaction finished, the rp md b database was changed outside
Yum.
After the transaction finished, the rp md b database was changed outside Yum.
The transaction failed to finish.
The transaction finished successfully, but yu m returned a non-zero exit code.
The transaction finished successfully, but an error or a warning was
displayed.
The transaction finished successfully, but problems already existed in the
rp md b database.
The transaction finished successfully, but the - - skip - b ro ken command-line
option was used and certain packages were skipped.

>
*
#
E
P
s

78

⁠Chapt er 7 . Yum

Yum also allows you to display a summary of all past transactions. To do so, run the command in
the following form as ro o t :
yu m h ist o ry su mmary
To display only transactions in a given range, type:
yu m h ist o ry su mmary start_id..end_id
Similarly to the yu m h ist o ry list command, you can also display a summary of transactions
regarding a certain package or packages by supplying a package name or a glob expression:
yu m h ist o ry su mmary glob_expression…
For instance, a summary of the transaction history displayed above would look like the following:
~]# yu m h ist o ry su mmary 1..5
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Login user
| Time
| Action(s)
| Altered
------------------------------------------------------------------------------Jaromir ...  | Last day
| Install
|
1
Jaromir ...  | Last week
| Install
|
1
Jaromir ...  | Last 2 weeks
| I, U
|
73
System 
| Last 2 weeks
| I, U
| 1107
history summary
All forms of the yu m h ist o ry su mmary command produce simplified tabular output similar to the
output of yu m h ist o ry list .
As shown above, both yu m h ist o ry list and yu m h ist o ry su mmary are oriented towards
transactions, and although they allow you to display only transactions related to a given package or
packages, they lack important details, such as package versions. To list transactions from the
perspective of a package, run the following command as ro o t :
yu m h ist o ry p ackag e- list glob_expression…
For example, to trace the history of subscription-manager and related packages, type the following at a
shell prompt:
~]# yu m h ist o ry p ackag e- list su b scrip t io n - man ag er\*
Loaded plugins: product-id, refresh-packagekit, subscription-manager
ID | Action(s)
| Package
------------------------------------------------------------------------------3 | Updated
| subscription-manager-0.95.11-1.el6.x86_64
3 | Update
|
0.95.17-1.el6_1.x86_64
3 | Updated
| subscription-manager-firstboot-0.95.11-1.el6.x86_64
3 | Update
|
0.95.17-1.el6_1.x86_64
3 | Updated
| subscription-manager-gnome-0.95.11-1.el6.x86_64
3 | Update
|
0.95.17-1.el6_1.x86_64
1 | Install
| subscription-manager-0.95.11-1.el6.x86_64
1 | Install
| subscription-manager-firstboot-0.95.11-1.el6.x86_64
1 | Install
| subscription-manager-gnome-0.95.11-1.el6.x86_64
history package-list

79

Deployment G uide

In this example, three packages were installed during the initial system installation: subscriptionmanager, subscription-manager-firstboot, and subscription-manager-gnome. In the third transaction, all
these packages were updated from version 0.95.11 to version 0.95.17.

7.3.2. Examining T ransact ions
To display the summary of a single transaction, as ro o t , use the yu m h ist o ry su mmary command
in the following form:
yu m h ist o ry su mmary id
To examine a particular transaction or transactions in more detail, run the following command as
ro o t :
yu m h ist o ry in f o id…
The id argument is optional and when you omit it, yu m automatically uses the last transaction. Note
that when specifying more than one transaction, you can also use a range:
yu m h ist o ry in f o start_id..end_id
The following is sample output for two transactions, each installing one new package:
~]# yu m h ist o ry in f o 4 ..5
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Transaction ID : 4..5
Begin time : Thu Jul 21 15:10:46 2011
Begin rpmdb : 1107:0c67c32219c199f92ed8da7572b4c6df64eacd3a
End time
:
15:33:15 2011 (22 minutes)
End rpmdb
: 1109:1171025bd9b6b5f8db30d063598f590f1c1f3242
User
: Jaromir Hradilek 
Return-Code : Success
Command Line : install screen
Command Line : install yum-plugin-security
Transaction performed with:
Installed rpm-4.8.0-16.el6.x86_64
Installed yum-3.2.29-17.el6.noarch
Installed yum-metadata-parser-1.1.2-16.el6.x86_64
Packages Altered:
Install screen-4.0.3-16.el6.x86_64
Install yum-plugin-security-1.1.30-17.el6.noarch
history info
You can also view additional information, such as what configuration options were used at the time
of the transaction, or from what repository and why were certain packages installed. To determine
what additional information is available for a certain transaction, type the following at a shell prompt
as ro o t :
yu m h ist o ry ad d o n - in f o id
Similarly to yu m h ist o ry in f o , when no id is provided, yu m automatically uses the latest
transaction. Another way to see the latest transaction is to use the last keyword:
yu m h ist o ry ad d o n - in f o last

80

⁠Chapt er 7 . Yum

For instance, for the first transaction in the previous example, the yu m h ist o ry ad d o n - in f o
command would provide the following output:
~]# yu m h ist o ry ad d o n - in f o 4
Loaded plugins: product-id, refresh-packagekit, subscription-manager
Transaction ID : 4
Available additional history information:
config-main
config-repos
saved_tx
history addon-info
In this example, three types of information are available:
co n f ig - main — global Yum options that were in use during the transaction. See Section 7.4.1,
“ Setting [main] Options” for information on how to change global options.
co n f ig - rep o s — options for individual Yum repositories. See Section 7.4.2, “ Setting [repository]
Options” for information on how to change options for individual repositories.
saved _t x — the data that can be used by the yu m lo ad - t ran sact io n command in order to
repeat the transaction on another machine (see below).
To display selected type of additional information, run the following command as ro o t :
yu m h ist o ry ad d o n - in f o id information

7.3.3. Revert ing and Repeat ing T ransact ions
Apart from reviewing the transaction history, the yu m h ist o ry command provides means to revert or
repeat a selected transaction. To revert a transaction, type the following at a shell prompt as ro o t :
yu m h ist o ry u n d o id
To repeat a particular transaction, as ro o t , run the following command:
yu m h ist o ry red o id
Both commands also accept the last keyword to undo or repeat the latest transaction.
Note that both yu m h ist o ry u n d o and yu m h ist o ry red o commands only revert or repeat the steps
that were performed during a transaction. If the transaction installed a new package, the yu m
h ist o ry u n d o command will uninstall it, and if the transaction uninstalled a package the command
will again install it. This command also attempts to downgrade all updated packages to their
previous version, if these older packages are still available.
When managing several identical systems, Yum also allows you to perform a transaction on one of
them, store the transaction details in a file, and after a period of testing, repeat the same transaction
on the remaining systems as well. To store the transaction details to a file, type the following at a
shell prompt as ro o t :
yu m - q h ist o ry ad d o n - in f o id saved _t x > f ile_n ame

81

Deployment G uide

Once you copy this file to the target system, you can repeat the transaction by using the following
command as ro o t :
yu m lo ad - t ran sact io n file_name
Note, however that the rp md b version stored in the file must be identical to the version on the target
system. You can verify the rp md b version by using the yu m versio n n o g ro u p s command.

7.3.4 . Complet ing T ransact ions
An unexpected situation, such as power loss or system crash, can prevent you from completing your
yum transaction. When such event occurs in the middle of your transaction, you can try to resume it
later with the following command as ro o t :
yu m- co mp let e- t ran sact io n
The yu m- co mp let e- t ran sact io n tool searches for incomplete or aborted yum transactions on a
system and attempts to complete them. By default, these transactions are listed in the
/var/lib /yu m/t ran sact io n - all and /var/lib /yu m/t ran sact io n - d o n e files. If there are more
unfinished transactions, yu m- co mp let e- t ran sact io n attempts to complete the most recent one
first.
To clean transaction journal files without attempting to resume the aborted transactions, use the - clean u p - o n ly option:
yu m- co mp let e- t ran sact io n - - clean u p - o n ly

7.3.5. St art ing New T ransact ion Hist ory
Yum stores the transaction history in a single SQLite database file. To start new transaction history,
run the following command as ro o t :
yu m h ist o ry n ew
This will create a new, empty database file in the /var/lib /yu m/h ist o ry/ directory. The old transaction
history will be kept, but will not be accessible as long as a newer database file is present in the
directory.

7.4 . Configuring Yum and Yum Reposit ories
The configuration file for yu m and related utilities is located at /et c/yu m.co n f . This file contains one
mandatory [ main ] section, which allows you to set Yum options that have global effect, and can
also contain one or more [ repository] sections, which allow you to set repository-specific options.
However, it is recommended to define individual repositories in new or existing .rep o files in the
/et c/yu m.rep o s.d /directory. The values you define in the [ main ] section of the /et c/yu m.co n f file
can override values set in individual [ repository] sections.
This section shows you how to:
set global Yum options by editing the [ main ] section of the /et c/yu m.co n f configuration file;
set options for individual repositories by editing the [ repository] sections in /et c/yu m.co n f and
.rep o files in the /et c/yu m.rep o s.d / directory;

82

⁠Chapt er 7 . Yum

use Yum variables in /et c/yu m.co n f and files in the /et c/yu m.rep o s.d / directory so that
dynamic version and architecture values are handled correctly;
add, enable, and disable Yum repositories on the command line; and,
set up your own custom Yum repository.

7.4 .1. Set t ing [main] Opt ions
The /et c/yu m.co n f configuration file contains exactly one [ main ] section, and while some of the
key-value pairs in this section affect how yu m operates, others affect how Yum treats repositories.
You can add many additional options under the [ main ] section heading in /et c/yu m.co n f .
A sample /et c/yu m.co n f configuration file can look like this:
[main]
cachedir=/var/cache/yum/$basearch/$releasever
keepcache=0
debuglevel=2
logfile=/var/log/yum.log
exactarch=1
obsoletes=1
gpgcheck=1
plugins=1
installonly_limit=3
[comments abridged]
# PUT YOUR REPOS HERE OR IN separate files named file.repo
# in /etc/yum.repos.d
The following are the most commonly-used options in the [ main ] section:
assu meyes= value
…where value is one of:
0 — yu m should prompt for confirmation of critical actions it performs. This is the default.
1 — D o not prompt for confirmation of critical yu m actions. If assu meyes= 1 is set, yu m
behaves in the same way that the command-line option - y does.
cach ed ir= directory
…where directory is an absolute path to the directory where Yum should store its cache and
database files. By default, Yum's cache directory is
/var/cach e/yu m/$b asearch /$releasever.
See Section 7.4.3, “ Using Yum Variables” for descriptions of the $b asearch and
$releasever Yum variables.
d eb u g level= value
…where value is an integer between 1 and 10. Setting a higher d eb u g level value causes
yu m to display more detailed debugging output. d eb u g level= 0 disables debugging
output, while d eb u g level= 2 is the default.
exact arch = value

83

Deployment G uide

…where value is one of:
0 — D o not take into account the exact architecture when updating packages.
1 — Consider the exact architecture when updating packages. With this setting, yu m will
not install an i686 package to update an i386 package already installed on the system.
This is the default.
exclu d e= package_name [ more_package_names]
This option allows you to exclude packages by keyword during installation/updates.
Listing multiple packages for exclusion can be accomplished by quoting a space-delimited
list of packages. Shell globs using wildcards (for example, * and ?) are allowed.
g p g ch eck= value
…where value is one of:
0 — D isable GPG signature-checking on packages in all repositories, including local
package installation.
1 — Enable GPG signature-checking on all packages in all repositories, including local
package installation. g p g ch eck= 1 is the default, and thus all packages' signatures are
checked.
If this option is set in the [ main ] section of the /et c/yu m.co n f file, it sets the GPG-checking
rule for all repositories. However, you can also set g p g ch eck= value for individual
repositories instead; that is, you can enable GPG-checking on one repository while
disabling it on another. Setting g p g ch eck= value for an individual repository in its
corresponding .rep o file overrides the default if it is present in /et c/yu m.co n f .
For more information on GPG signature-checking, see Section B.3, “ Checking a Package's
Signature” .
g ro u p remo ve_leaf _o n ly= value
…where value is one of:
0 — yu m should not check the dependencies of each package when removing a package
group. With this setting, yu m removes all packages in a package group, regardless of
whether those packages are required by other packages or groups.
g ro u p remo ve_leaf _o n ly= 0 is the default.
1 — yu m should check the dependencies of each package when removing a package
group, and remove only those packages which are not required by any other package or
group.
For more information on removing packages, see Intelligent package group removal.
in st allo n lyp kg s= space separated list of packages
Here you can provide a space-separated list of packages which yu m can install, but will
never update. See the yu m.co n f (5) manual page for the list of packages which are installonly by default.
If you add the in st allo n lyp kg s directive to /et c/yu m.co n f , you should ensure that you
list all of the packages that should be install-only, including any of those listed under the
in st allo n lyp kg s section of yu m.co n f (5). In particular, kernel packages should always
be listed in in st allo n lyp kg s (as they are by default), and in st allo n ly_limit should

84

⁠Chapt er 7 . Yum

always be set to a value greater than 2 so that a backup kernel is always available in case
the default one fails to boot.
⁠i n st allo n ly_limit = value
…where value is an integer representing the maximum number of versions that can be
installed simultaneously for any single package listed in the in st allo n lyp kg s directive.
The defaults for the in st allo n lyp kg s directive include several different kernel packages,
so be aware that changing the value of in st allo n ly_limit will also affect the maximum
number of installed versions of any single kernel package. The default value listed in
/et c/yu m.co n f is in st allo n ly_limit = 3, and it is not recommended to decrease this value,
particularly below 2.
keep cach e= value
…where value is one of:
0 — D o not retain the cache of headers and packages after a successful installation. This
is the default.
1 — Retain the cache after a successful installation.
lo g f ile= file_name
…where file_name is an absolute path to the file in which yu m should write its logging
output. By default, yu m logs to /var/lo g /yu m.lo g .
mu lt ilib _p o licy= value
…where value is one of:
b est — install the best-choice architecture for this system. For example, setting
mu lt ilib _p o licy= b est on an AMD 64 system causes yu m to install 64-bit versions of all
packages.
all — always install every possible architecture for every package. For example, with
mu lt ilib _p o licy set to all on an AMD 64 system, yu m would install both the i686 and
AMD 64 versions of a package, if both were available.
o b so let es= value
…where value is one of:
0 — D isable yu m's obsoletes processing logic when performing updates.
1 — Enable yu m's obsoletes processing logic when performing updates. When one
package declares in its spec file that it obsoletes another package, the latter package will be
replaced by the former package when the former package is installed. Obsoletes are
declared, for example, when a package is renamed. o b so let es= 1 the default.
p lu g in s= value
…where value is one of:
0 — D isable all Yum plug-ins globally.

85

Deployment G uide

Disabling all plug-ins is not advised
D isabling all plug-ins is not advised because certain plug-ins provide important
Yu m services. In particular, rh n p lu g in provides support for R H N C lassic, and
p ro d u ct - id and su b scrip t io n - man ag er plug-ins provide support for the
certificate-based C o n t en t D elivery N et wo rk (CD N). D isabling plug-ins globally is
provided as a convenience option, and is generally only recommended when
diagnosing a potential problem with Yu m.
1 — Enable all Yum plug-ins globally. With p lu g in s= 1, you can still disable a specific Yum
plug-in by setting en ab led = 0 in that plug-in's configuration file.
For more information about various Yum plug-ins, see Section 7.5, “ Yum Plug-ins” . For
further information on controlling plug-ins, see Section 7.5.1, “ Enabling, Configuring, and
D isabling Yum Plug-ins” .
rep o sd ir= directory
…where directory is an absolute path to the directory where .rep o files are located. All
.rep o files contain repository information (similar to the [ repository] sections of
/et c/yu m.co n f ). yu m collects all repository information from .rep o files and the
[ repository] section of the /et c/yu m.co n f file to create a master list of repositories to use
for transactions. If rep o sd ir is not set, yu m uses the default directory /et c/yu m.rep o s.d /.
ret ries= value
…where value is an integer 0 or greater. This value sets the number of times yu m should
attempt to retrieve a file before returning an error. Setting this to 0 makes yu m retry forever.
The default value is 10.
For a complete list of available [ main ] options, see the [ main ] O PT IO N S section of the
yu m.co n f (5) manual page.

7.4 .2. Set t ing [reposit ory] Opt ions
The [ repository] sections, where repository is a unique repository ID such as my_p erso n al_rep o
(spaces are not permitted), allow you to define individual Yum repositories. To avoid conflicts,
custom repositories should not use names used by Red Hat repositories.
The following is a bare-minimum example of the form a [ repository] section takes:
[repository]
name=repository_name
baseurl=repository_url
Every [ repository] section must contain the following directives:
n ame= repository_name
…where repository_name is a human-readable string describing the repository.
b aseu rl= repository_url
…where repository_url is a URL to the directory where the repodata directory of a repository
is located:

86

⁠Chapt er 7 . Yum

If the repository is available over HTTP, use: h t t p ://p at h /t o /rep o
If the repository is available over FTP, use: f t p ://p at h /t o /rep o
If the repository is local to the machine, use: f ile:///p at h /t o /lo cal/rep o
If a specific online repository requires basic HTTP authentication, you can specify your
user name and password by prepending it to the URL as username:password@ link. For
example, if a repository on http://www.example.com/repo/ requires a user name of “ user”
and a password of “ password” , then the b aseu rl link could be specified as
h t t p ://u ser:p asswo rd @ www.examp le.co m/rep o /.
Usually this URL is an HTTP link, such as:
baseurl=http://path/to/repo/releases/$releasever/server/$basearch/os/
Note that Yum always expands the $releasever, $arch , and $b asearch variables in
URLs. For more information about Yum variables, see Section 7.4.3, “ Using Yum Variables” .
Another useful [ repository] directive is the following:
en ab led = value
…where value is one of:
0 — D o not include this repository as a package source when performing updates and
installs. This is an easy way of quickly turning repositories on and off, which is useful when
you desire a single package from a repository that you do not want to enable for updates or
installs.
1 — Include this repository as a package source.
Turning repositories on and off can also be performed by passing either the - en ab lerep o = repo_name or - - d isab lerep o = repo_name option to yu m, or through the
Ad d /R emo ve So f t ware window of the Packag eK it utility.
Many more [ repository] options exist. For a complete list, see the [ rep o sit o ry] O PT IO N S section of
the yu m.co n f (5) manual page.

Examp le 7.5. A samp le /et c/yu m.rep o s.d /red h at .rep o f ile
The following is a sample /et c/yu m.rep o s.d /red h at .rep o file:
#
# Red Hat Repositories
# Managed by (rhsm) subscription-manager
#
[red-hat-enterprise-linux-scalable-file-system-for-rhel-6-entitlement-rpms]
name = Red Hat Enterprise Linux Scalable File System (for RHEL 6 Entitlement) (RPMs)
baseurl = https://cdn.redhat.com/content/dist/rhel/entitlement6/releases/$releasever/$basearch/scalablefilesystem/os
enabled = 1
gpgcheck = 1
gpgkey = file:///etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release
sslverify = 1
sslcacert = /etc/rhsm/ca/redhat-uep.pem

87

Deployment G uide

sslclientkey = /etc/pki/entitlement/key.pem
sslclientcert = /etc/pki/entitlement/11300387955690106.pem
[red-hat-enterprise-linux-scalable-file-system-for-rhel-6-entitlement-source-rpms]
name = Red Hat Enterprise Linux Scalable File System (for RHEL 6 Entitlement) (Source
RPMs)
baseurl = https://cdn.redhat.com/content/dist/rhel/entitlement6/releases/$releasever/$basearch/scalablefilesystem/source/SRPMS
enabled = 0
gpgcheck = 1
gpgkey = file:///etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release
sslverify = 1
sslcacert = /etc/rhsm/ca/redhat-uep.pem
sslclientkey = /etc/pki/entitlement/key.pem
sslclientcert = /etc/pki/entitlement/11300387955690106.pem
[red-hat-enterprise-linux-scalable-file-system-for-rhel-6-entitlement-debug-rpms]
name = Red Hat Enterprise Linux Scalable File System (for RHEL 6 Entitlement) (D ebug RPMs)
baseurl = https://cdn.redhat.com/content/dist/rhel/entitlement6/releases/$releasever/$basearch/scalablefilesystem/debug
enabled = 0
gpgcheck = 1
gpgkey = file:///etc/pki/rpm-gpg/RPM-GPG-KEY-redhat-release
sslverify = 1
sslcacert = /etc/rhsm/ca/redhat-uep.pem
sslclientkey = /etc/pki/entitlement/key.pem
sslclientcert = /etc/pki/entitlement/11300387955690106.pem

7.4 .3. Using Yum Variables
You can use and reference the following built-in variables in yu m commands and in all Yum
configuration files (that is, /et c/yu m.co n f and all .rep o files in the /et c/yu m.rep o s.d / directory):
$releasever
You can use this variable to reference the release version of Red Hat Enterprise Linux. Yum
obtains the value of $releasever from the d ist ro verp kg = value line in the /et c/yu m.co n f
configuration file. If there is no such line in /et c/yu m.co n f , then yu m infers the correct
value by deriving the version number from the redhat-release package. The value of
$releasever typically consists of the major release number and the variant of Red Hat
Enterprise Linux, for example 6 C lien t , or 6 Server.
$arch
You can use this variable to refer to the system's CPU architecture as returned when calling
Python's o s.u n ame( ) function. Valid values for $arch include i6 86 and x86 _6 4 .
$b asearch
You can use $b asearch to reference the base architecture of the system. For example, i686
machines have a base architecture of i386 , and AMD 64 and Intel64 machines have a base
architecture of x86 _6 4 .
$YU M0- 9

88

⁠Chapt er 7 . Yum

These ten variables are each replaced with the value of any shell environment variables
with the same name. If one of these variables is referenced (in /et c/yu m.co n f for example)
and a shell environment variable with the same name does not exist, then the configuration
file variable is not replaced.
To define a custom variable or to override the value of an existing one, create a file with the same
name as the variable (without the “ $” sign) in the /et c/yu m/vars/ directory, and add the desired
value on its first line.
For example, repository descriptions often include the operating system name. To define a new
variable called $o sn ame, create a new file with “ Red Hat Enterprise Linux” on the first line and save
it as /et c/yu m/vars/o sn ame:
~]# ech o "R ed H at En t erp rise Lin u x" > /et c/yu m/vars/o sn ame
Instead of “ Red Hat Enterprise Linux 6” , you can now use the following in the .rep o files:
name=$osname $releasever

7.4 .4 . Viewing t he Current Configurat ion
To display the current values of global Yum options (that is, the options specified in the [ main ]
section of the /et c/yu m.co n f file), run the yu m- co n f ig - man ag er with no command-line options:
yu m- co n f ig - man ag er
To list the content of a different configuration section or sections, use the command in the following
form:
yu m- co n f ig - man ag er section…
You can also use a glob expression to display the configuration of all matching sections:
yu m- co n f ig - man ag er glob_expression…
For example, to list all configuration options and their corresponding values, type the following at a
shell prompt:
~]$ yu m- co n f ig - man ag er main \*
Loaded plugins: product-id, refresh-packagekit, subscription-manager
================================== main ===================================
[main]
alwaysprompt = True
assumeyes = False
bandwith = 0
bugtracker_url = https://bugzilla.redhat.com/enter_bug.cgi?
product=Red% 20Hat% 20Enterprise% 20Linux% 206&component=yum
cache = 0
[output truncated]

7.4 .5. Adding, Enabling, and Disabling a Yum Reposit ory

89

Deployment G uide

Section 7.4.2, “ Setting [repository] Options” described various options you can use to define a Yum
repository. This section explains how to add, enable, and disable a repository by using the yu mco n f ig - man ag er command.

T he /etc/yum.repos.d/redhat.repo file
When the system is registered with the certificate-based R ed H at N et wo rk, the Red Hat
Subscription Manager tools are used to manage repositories in the
/et c/yu m.rep o s.d /red h at .rep o file. See Chapter 5, Registering the System and Managing
Subscriptions for more information how to register a system with R ed H at N et wo rk and use
the Red Hat Subscription Manager tools to manage subscriptions.

Adding a Yum Re po sit o ry
To define a new repository, you can either add a [ repository] section to the /et c/yu m.co n f file, or to
a .rep o file in the /et c/yu m.rep o s.d / directory. All files with the .rep o file extension in this directory
are read by yu m, and it is recommended to define your repositories here instead of in
/et c/yu m.co n f .

Be careful when using untrusted software sources
Obtaining and installing software packages from unverified or untrusted software sources
other than Red Hat Network constitutes a potential security risk, and could lead to security,
stability, compatibility, and maintainability issues.
Yum repositories commonly provide their own .rep o file. To add such a repository to your system
and enable it, run the following command as ro o t :
yu m- co n f ig - man ag er - - ad d - rep o repository_url
…where repository_url is a link to the .rep o file. For example, to add a repository located at
http://www.example.com/example.repo, type the following at a shell prompt:
~]# yu m- co n f ig - man ag er - - ad d - rep o h t t p ://www.examp le.co m/examp le.rep o
Loaded plugins: product-id, refresh-packagekit, subscription-manager
adding repo from: http://www.example.com/example.repo
grabbing file http://www.example.com/example.repo to /etc/yum.repos.d/example.repo
example.repo
| 413 B 00:00
repo saved to /etc/yum.repos.d/example.repo

Enabling a Yum Re po sit o ry
To enable a particular repository or repositories, type the following at a shell prompt as ro o t :
yu m- co n f ig - man ag er - - en ab le repository…
…where repository is the unique repository ID (use yu m rep o list all to list available repository ID s).
Alternatively, you can use a glob expression to enable all matching repositories:
yu m- co n f ig - man ag er - - en ab le glob_expression…

90

⁠Chapt er 7 . Yum

For example, to enable repositories defined in the [ examp le] , [ examp le- d eb u g in f o ] , and
[ examp le- so u rce] sections, type:
~]# yu m- co n f ig - man ag er - - en ab le examp le\*
Loaded plugins: product-id, refresh-packagekit, subscription-manager
============================== repo: example ==============================
[example]
bandwidth = 0
base_persistdir = /var/lib/yum/repos/x86_64/6Server
baseurl = http://www.example.com/repo/6Server/x86_64/
cache = 0
cachedir = /var/cache/yum/x86_64/6Server/example
[output truncated]
When successful, the yu m- co n f ig - man ag er - - en ab le command displays the current repository
configuration.

Disabling a Yum Re po sit o ry
To disable a Yum repository, run the following command as ro o t :
yu m- co n f ig - man ag er - - d isab le repository…
…where repository is the unique repository ID (use yu m rep o list all to list available repository ID s).
Similarly to yu m- co n f ig - man ag er - - en ab le, you can use a glob expression to disable all
matching repositories at the same time:
yu m- co n f ig - man ag er - - d isab le glob_expression…
When successful, the yu m- co n f ig - man ag er - - d isab le command displays the current
configuration.

7.4 .6. Creat ing a Yum Reposit ory
To set up a Yum repository, follow these steps:
1. Install the createrepo package. To do so, type the following at a shell prompt as ro o t :
yu m in st all creat erep o
2. Copy all packages that you want to have in your repository into one directory, such as
/mn t /lo cal_rep o /.
3. Change to this directory and run the following command:
creat erep o - - d at ab ase /mn t /lo cal_rep o
This creates the necessary metadata for your Yum repository, as well as the sq lit e database
for speeding up yu m operations.

91

Deployment G uide

Using the createrepo command on Red Hat Enterprise Linux
5
Compared to Red Hat Enterprise Linux 5, RPM packages for Red Hat
Enterprise Linux 6 are compressed with the XZ lossless data compression format and
can be signed with newer hash algorithms like SHA-256. Consequently, it is not
recommended to use the creat erep o command on Red Hat Enterprise Linux 5 to
create the package metadata for Red Hat Enterprise Linux 6.

7.4 .7. Working wit h Yum Cache
By default, yu m deletes downloaded data files when they are no longer needed after a successful
operation. This minimizes the amount of storage space that yu m uses. However, you can enable
caching, so that the package files downloaded by yu m stay in cache directories. By using cached
data, you can carry out certain operations without a network connection, you can also copy
packages stored in the caches and reuse them elsewhere.
Yum stores temporary files in the /var/cach e/yu m/$b asearch /$releasever/ directory, where
$b asearch and $releasever are Yum variables referring to base architecture of the system and the
release version of Red Hat Enterprise Linux. Each configured repository has one subdirectory. For
example, the directory /var/cach e/yu m/$b asearch /$releasever/d evelo p men t /p ackag es/ holds
packages downloaded from the development repository. You can find the values for the $basearch
and $releasever variables in the output of the yu m versio n command.
To change the default cache location, modify the cach ed ir option in the [ main ] section of the
/et c/yu m.co n f configuration file. See Section 7.4, “ Configuring Yum and Yum Repositories” for more
information on configuring yu m.

Enabling t he Cache s
To retain the cache of packages after a successful installation, add the following text to the [ main ]
section of /et c/yu m.co n f .
keep cach e = 1
Once you enabled caching, every yum operation may download package data from the configured
repositories.
To download and make usable all the metadata for the currently enabled yum repositories, type:
yu m makecach e
This is useful if you want to make sure that the cache is fully up to date with all metadata. To set the
time after which the metadata will expire, use the met ad at a- exp ire setting in /et c/yu m.co n f .

Using yum in Cache -o nly Mo de
To carry out a yu m command without a network connection, add the - C or - - cach eo n ly commandline option. With this option, yu m proceeds without checking any network repositories, and uses only
cached files. In this mode, yu m may only install packages that have been downloaded and cached
by a previous operation.

92

⁠Chapt er 7 . Yum

For instance, to list packages that use the currently cached data with names that contain
“ gstreamer” , enter the following command:
yu m - C list gstreamer*

Cle aring t he yum Cache s
It is often useful to remove entries accumulated in the /var/cach e/yu m/ directory. If you remove a
package from the cache, you do not affect the copy of the software installed on your system. To
remove all entries for currently enabled repositories from the cache, type the following as a ro o t :
yu m clean all
There are various ways to invoke yu m in clean mode depending on the type of cached data you
want to remove. See Table 7.3, “ Available yu m clean options” for a complete list of available
configuration options.
T ab le 7.3. Availab le yu m clean o p t io n s
O p t io n

D escrip t io n

expire-cache

eliminates time records of the metadata and
mirrorlists download for each repository. This
forces yum to revalidate the cache for each
repository the next time it is used.
eliminates any cached packages from the
system
eliminates all header files that previous versions
of yum used for dependency resolution
eliminates all files that yum uses to determine the
remote availability of packages. These metadata
are downloaded again the next time yum is run.
eliminates the sqlite cache used for faster
access to metadata. Using this option will force
yum to download the sqlite metadata the next
time it is run. This does not apply for
repositories that contain only .xml data, in that
case, sqlite data are deleted but without
subsequent download
eliminates any cached data from the local
rpmdb
enabled plugins are forced to eliminate their
cached data
removes all of the above

packages
headers
metadata

dbcache

rpmdb
plugins
all

The exp ire- cach e option is most preferred from the above list. In many cases, it is a sufficient and
much faster replacement for clean all.

7.4 .8. Adding t he Opt ional and Supplement ary Reposit ories
Optional and Supplementary subscription channels provide additional software packages for
Red Hat Enterprise Linux that cover open source licensed software (in the Optional channel) and
proprietary licensed software (in the Supplementary channel).

93

Deployment G uide

Before subscribing to the Optional and Supplementary channels see the Scope of Coverage D etails.
If you decide to install packages from these channels, follow the steps documented in the article
called How to access Optional and Supplementary channels, and -devel packages using Red Hat
Subscription Manager (RHSM)? on the Red Hat Customer Portal.

7.5. Yum Plug-ins
Yum provides plug-ins that extend and enhance its operations. Certain plug-ins are installed by
default. Yum always informs you which plug-ins, if any, are loaded and active whenever you call any
yu m command. For example:
~]# yu m in f o yu m
Loaded plugins: product-id, refresh-packagekit, subscription-manager
[output truncated]
Note that the plug-in names which follow Lo ad ed p lu g in s are the names you can provide to the - d isab lep lu g in s= plugin_name option.

7.5.1. Enabling, Configuring, and Disabling Yum Plug-ins
To enable Yum plug-ins, ensure that a line beginning with p lu g in s= is present in the [ main ] section
of /et c/yu m.co n f , and that its value is 1:
plugins=1
You can disable all plug-ins by changing this line to p lu g in s= 0.

Disabling all plug-ins is not advised
D isabling all plug-ins is not advised because certain plug-ins provide important Yu m
services. In particular, rh n p lu g in provides support for R H N C lassic, and p ro d u ct - id and
su b scrip t io n - man ag er plug-ins provide support for the certificate-based C o n t en t
D elivery N et wo rk (CD N). D isabling plug-ins globally is provided as a convenience option,
and is generally only recommended when diagnosing a potential problem with Yu m.
Every installed plug-in has its own configuration file in the /et c/yu m/p lu g in co n f .d / directory. You
can set plug-in specific options in these files. For example, here is the ref resh - p ackag ekit plugin's ref resh - p ackag ekit .co n f configuration file:
[main]
enabled=1
Plug-in configuration files always contain a [ main ] section (similar to Yum's /et c/yu m.co n f file) in
which there is (or you can place if it is missing) an en ab led = option that controls whether the plugin is enabled when you run yu m commands.
If you disable all plug-ins by setting en ab led = 0 in /et c/yu m.co n f , then all plug-ins are disabled
regardless of whether they are enabled in their individual configuration files.
If you merely want to disable all Yum plug-ins for a single yu m command, use the - - n o p lu g in s
option.

94

⁠Chapt er 7 . Yum

If you want to disable one or more Yum plug-ins for a single yu m command, add the - d isab lep lu g in = plugin_name option to the command. For example, to disable the p rest o plug-in
while updating a system, type:
~]# yu m u p d at e - - d isab lep lu g in = p rest o
The plug-in names you provide to the - - d isab lep lu g in = option are the same names listed after the
Lo ad ed p lu g in s line in the output of any yu m command. You can disable multiple plug-ins by
separating their names with commas. In addition, you can match multiple plug-in names or shorten
long ones by using glob expressions:
~]# yu m u p d at e - - d isab lep lu g in = p rest o ,ref resh - p ack*

7.5.2. Inst alling Addit ional Yum Plug-ins
Yum plug-ins usually adhere to the yu m- p lu g in - plugin_name package-naming convention, but not
always: the package which provides the p rest o plug-in is named yu m- p rest o , for example. You
can install a Yum plug-in in the same way you install other packages. For instance, to install the
secu rit y plug-in, type the following at a shell prompt:
~]# yu m in st all yu m- p lu g in - secu rit y

7.5.3. Plug-in Descript ions
The following list provides descriptions of a few useful Yum plug-ins:
kab i ( kabi-yum-plugins)
The kab i plug-in checks whether a driver update package conforms with official Red Hat
kernel Application Binary Interface (kABI). With this plug-in enabled, when a user attempts to
install a package that uses kernel symbols which are not on a whitelist, a warning message
is written to the system log. Additionally, configuring the plug-in to run in enforcing mode
prevents such packages from being installed at all.
To configure the kab i plug-in, edit the configuration file located in
/et c/yu m/p lu g in co n f .d /kab i.co n f . See Table 7.4, “ Supported kab i.co n f directives” for
a list of directives that can be used in the [ main ] section.
T ab le 7.4 . Su p p o rt ed kab i.co n f d irect ives
D irect ive

D escrip t io n

en ab led =value

Allows you to enable or disable the plug-in. The value must be
either 1 (enabled), or 0 (disabled). When installed, the plug-in
is enabled by default.
Allows you to specify the directory in which the files with
supported kernel symbols are located. By default, the kab i
plug-in uses files provided by the kernel-abi-whitelists package
(that is, the /lib /mo d u les/kab i/ directory).
Allows you to enable or disable enforcing mode. The value
must be either 1 (enabled), or 0 (disabled). By default, this
option is commented out and the kab i plug-in only displays a
warning message.

wh it elist s=directory

en f o rce=value

p rest o ( yum-presto)

95

Deployment G uide

The p rest o plug-in adds support to Yum for downloading delta RPM packages, during
updates, from repositories which have p rest o metadata enabled. D elta RPMs contain only
the differences between the version of the package installed on the client requesting the
RPM package and the updated version in the repository.
D ownloading a delta RPM is much quicker than downloading the entire updated package,
and can speed up updates considerably. Once the delta RPMs are downloaded, they must
be rebuilt to apply the difference to the currently-installed package and thus create the full,
updated package. This process takes CPU time on the installing machine. Using delta
RPMs is therefore a compromise between time-to-download, which depends on the network
connection, and time-to-rebuild, which is CPU-bound. Using the p rest o plug-in is
recommended for fast machines and systems with slower network connections, while slower
machines on very fast connections benefit more from downloading normal RPM packages,
that is, by disabling p rest o .
p ro d u ct - id ( subscription-manager)
The p ro d u ct - id plug-in manages product identity certificates for products installed from
the Content D elivery Network. The p ro d u ct - id plug-in is installed by default.
ref resh - p ackag ekit ( PackageKit-yum-plugin)
The ref resh - p ackag ekit plug-in updates metadata for Packag eK it whenever yu m is
run. The ref resh - p ackag ekit plug-in is installed by default.
rh n p lu g in ( yum-rhn-plugin)
The rh n p lu g in provides support for connecting to R H N C lassic. This allows systems
registered with R H N C lassic to update and install packages from this system. Note that
R H N C lassic is only provided for older Red Hat Enterprise Linux systems (that is, Red Hat
Enterprise Linux 4.x, Red Hat Enterprise Linux 5.x, and Satellite 5.x) in order to migrate them
over to Red Hat Enterprise Linux 6. The rh n p lu g in is installed by default.
See the rh n p lu g in (8) manual page for more information about the plug-in.
secu rit y ( yum-plugin-security)
D iscovering information about and applying security updates easily and often is important
to all system administrators. For this reason Yum provides the secu rit y plug-in, which
extends yu m with a set of highly-useful security-related commands, subcommands and
options.
You can check for security-related updates as follows:
~]# yu m ch eck- u p d at e - - secu rit y
Loaded plugins: product-id, refresh-packagekit, security, subscription-manager
Updating Red Hat repositories.
INFO:rhsm-app.repolib:repos updated: 0
Limiting package lists to security relevant ones
Needed 3 of 7 packages, for security
elinks.x86_64
0.12-0.13.el6
rhel
kernel.x86_64
2.6.30.8-64.el6
rhel
kernel-headers.x86_64
2.6.30.8-64.el6
rhel
You can then use either yu m u p d at e - - secu rit y or yu m u p d at e- min imal - - secu rit y to
update those packages which are affected by security advisories. Both of these commands
update all packages on the system for which a security advisory has been issued. yu m
u p d at e- min imal - - secu rit y updates them to the latest packages which were released as

96

⁠Chapt er 7 . Yum

part of a security advisory, while yu m u p d at e - - secu rit y will update all packages affected
by a security advisory to the latest version of that package available.
In other words, if:
the kernel-2.6.30.8-16 package is installed on your system;
the kernel-2.6.30.8-32 package was released as a security update;
then kernel-2.6.30.8-64 was released as a bug fix update,
...then yu m u p d at e- min imal - - secu rit y will update you to kernel-2.6.30.8-32, and yu m
u p d at e - - secu rit y will update you to kernel-2.6.30.8-64. Conservative system
administrators probably want to use u p d at e- min imal to reduce the risk incurred by
updating packages as much as possible.
See the yu m- secu rit y(8) manual page for usage details and further explanation of the
enhancements the secu rit y plug-in adds to yu m.
su b scrip t io n - man ag er ( subscription-manager)
The su b scrip t io n - man ag er plug-in provides support for connecting to R ed H at
N et wo rk. This allows systems registered with R ed H at N et wo rk to update and install
packages from the certificate-based Content D elivery Network. The su b scrip t io n man ag er plug-in is installed by default.
See Chapter 5, Registering the System and Managing Subscriptions for more information how to
manage product subscriptions and entitlements.
yu m- d o wn lo ad o n ly ( yum-plugin-downloadonly)
The yu m- d o wn lo ad o n ly plug-in provides the - - d o wn lo ad o n ly command-line option
which can be used to download packages from Red Hat Network or a configured Yum
repository without installing the packages.
To install the package, follow the instructions in Section 7.5.2, “ Installing Additional Yum
Plug-ins” . After the installation, see the contents of the
/et c/yu m/p lu g in co n f .d /d o wn lo ad o n ly.co n f file to ensure that the plug-in is enabled:
~]$ cat /et c/yu m/p lu g in co n f .d /d o wn lo ad o n ly.co n f
[main]
enabled=1
In the following example, the yu m in st all - - d o wn lo ad o n ly command is run to download
the latest version of the httpd package, without installing it:
~]# yu m in st all h t t p d - - d o wn lo ad o n ly
Loaded plugins: downloadonly, product-id, refresh-packagekit, rhnplugin,
: subscription-manager
Updating Red Hat repositories.
Setting up Install Process
Resolving D ependencies
--> Running transaction check
---> Package httpd.x86_64 0:2.2.15-9.el6_1.2 will be updated
---> Package httpd.x86_64 0:2.2.15-15.el6_2.1 will be an update
--> Processing D ependency: httpd-tools = 2.2.15-15.el6_2.1 for package: httpd-2.2.1515.el6_2.1.x86_64
--> Running transaction check

97

Deployment G uide

---> Package httpd-tools.x86_64 0:2.2.15-9.el6_1.2 will be updated
---> Package httpd-tools.x86_64 0:2.2.15-15.el6_2.1 will be an update
--> Finished D ependency Resolution
D ependencies Resolved
============================================================================
====
Package
Arch
Version
Repository
Size
============================================================================
====
Updating:
httpd
x86_64 2.2.15-15.el6_2.1
rhel-x86_64-server-6 812 k
Updating for dependencies:
httpd-tools x86_64 2.2.15-15.el6_2.1
rhel-x86_64-server-6 70 k
Transaction Summary
============================================================================
====
Upgrade
2 Package(s)
Total download size: 882 k
Is this ok [y/N]: y
D ownloading Packages:
(1/2): httpd-2.2.15-15.el6_2.1.x86_64.rpm
| 812 kB 00:00
(2/2): httpd-tools-2.2.15-15.el6_2.1.x86_64.rpm
| 70 kB 00:00
-------------------------------------------------------------------------------Total
301 kB/s | 882 kB 00:02

exiting because --downloadonly specified
By default, packages downloaded using the - - d o wn lo ad o n ly option are saved in one of
the subdirectories of the
⁠/var/cach e/yu m directory, depending on the Red Hat Enterprise Linux variant and
architecture.
If you want to specify an alternate directory to save the packages, pass the - - d o wn lo ad d ir
option along with - - d o wn lo ad o n ly:
~]# yu m in st all - - d o wn lo ad o n ly - - d o wn lo ad d ir= /p at h /t o /d irect o ry h t t p d

Note
As an alternative to the yu m- d o wn lo ad o n ly plugin — to download packages
without installing them — you can use the yu md o wn lo ad er utility that is provided
by the yum-utils package.

7.6. Addit ional Resources
For more information on how to manage software packages on Red Hat Enterprise Linux, see the
resources listed below.

98

⁠Chapt er 7 . Yum

Inst alled Document at ion
yu m(8) — The manual page for the yu m command-line utility provides a complete list of
supported options and commands.
yu md b (8) — The manual page for the yu md b command-line utility documents how to use this
tool to query and, if necessary, alter the yum database.
yu m.co n f (5) — The manual page named yu m.co n f documents available yum configuration
options.
yu m- u t ils(1) — The manual page named yu m- u t ils lists and briefly describes additional utilities
for managing yum configuration, manipulating repositories, and working with yum database.

Online Resources
Yum Guides — The Yum Guides page on the project home page provides links to further
documentation.
Red Hat Access Labs — The Red Hat Access Labs includes a “ Yum Repository Configuration
Helper” .

See Also
Chapter 4, Gaining Privileges documents how to gain administrative privileges by using the su and
su d o commands.
Appendix B, RPM describes the R PM Packag e Man ag er (RPM), the packaging system used by
Red Hat Enterprise Linux.

99

Deployment G uide

Chapter 8. PackageKit
Red Hat provides Packag eK it for viewing, managing, updating, installing and uninstalling
packages compatible with your system. PackageKit consists of several graphical interfaces that can
be opened from the GNOME panel menu, or from the Notification Area when PackageKit alerts you
that updates are available. For more information on Packag eK it ' s architecture and available front
ends, see Section 8.3, “ PackageKit Architecture” .

8.1. Updat ing Packages wit h Soft ware Updat e
PackageKit displays a starburst icon in the Notification Area whenever updates are available to be
installed on your system.

Fig u re 8.1. Packag eK it ' s ico n in t h e N o t if icat io n Area
Clicking on the notification icon opens the So f t ware U p d at e window. Alternatively, you can open
So f t ware U p d at es by clicking Syst em → Ad min ist rat io n → So f t ware U p d at e from the GNOME
panel, or running the g p k- u p d at e- viewer command at the shell prompt. In the So f t ware U p d at es
window, all available updates are listed along with the names of the packages being updated (minus
the .rp m suffix, but including the CPU architecture), a short summary of the package, and, usually,
short descriptions of the changes the update provides. Any updates you do not wish to install can be
de-selected here by unchecking the checkbox corresponding to the update.

100

⁠Chapt er 8 . PackageKit

Fig u re 8.2. In st allin g u p d at es wit h So f t ware U p d at e
The updates presented in the So f t ware U p d at es window only represent the currently-installed
packages on your system for which updates are available; dependencies of those packages, whether
they are existing packages on your system or new ones, are not shown until you click In st all
U p d at es.
PackageKit utilizes the fine-grained user authentication capabilities provided by the Po licyK it
toolkit whenever you request it to make changes to the system. Whenever you instruct PackageKit to
update, install or remove packages, you will be prompted to enter the superuser password before
changes are made to the system.
If you instruct PackageKit to update the kernel package, then it will prompt you after installation,
asking you whether you want to reboot the system and thereby boot into the newly-installed kernel.

Set t ing t he Updat e-Checking Int erval
Right-clicking on PackageKit's Notification Area icon and clicking Pref eren ces opens the
So f t ware U p d at e Pref eren ces window, where you can define the interval at which PackageKit
checks for package updates, as well as whether or not to automatically install all updates or only
security updates. Leaving the C h eck f o r u p d at es wh en u sin g mo b ile b ro ad b an d box
unchecked is handy for avoiding extraneous bandwidth usage when using a wireless connection on
which you are charged for the amount of data you download.

101

Deployment G uide

Fig u re 8.3. Set t in g Packag eK it ' s u p d at e- ch eckin g in t erval

8.2. Using Add/Remove Soft ware
To find and install a new package, on the GNOME panel click on Syst em → Ad min ist rat io n →
Ad d /R emo ve So f t ware, or run the g p k- ap p licat io n command at the shell prompt.

Fig u re 8.4 . Packag eK it ' s Ad d /R emo ve So f t ware win d o w

8.2.1. Refreshing Soft ware Sources (Yum Reposit ories)
PackageKit refers to Yu m repositories as software sources. It obtains all packages from enabled
software sources. You can view the list of all configured and unfiltered (see below) Yum repositories by
opening Ad d /R emo ve So f t ware and clicking Syst em → So f t ware so u rces. The So f t ware
So u rces dialog shows the repository name, as written on the n ame=  field of
all [repository] sections in the /et c/yu m.co n f configuration file, and in all repository.rep o files in the
/et c/yu m.rep o s.d / directory.
Entries which are checked in the En ab led column indicate that the corresponding repository will be
used to locate packages to satisfy all update and installation requests (including dependency
resolution). You can enable or disable any of the listed Yum repositories by selecting or clearing the
checkbox. Note that doing so causes Po licyK it to prompt you for superuser authentication.

102

⁠Chapt er 8 . PackageKit

The En ab led column corresponds to the en ab led = <1 or 0> field in [repository] sections. When you
click the checkbox, PackageKit inserts the en ab led = <1 or 0> line into the correct [repository] section
if it does not exist, or changes the value if it does. This means that enabling or disabling a repository
through the So f t ware So u rces window causes that change to persist after closing the window or
rebooting the system.
Note that it is not possible to add or remove Yum repositories through PackageKit.

Showing source RPM, test and debuginfo repositories
Checking the box at the bottom of the So f t ware So u rces window causes PackageKit to
display source RPM, testing and debuginfo repositories as well. This box is unchecked by
default.
After making a change to the available Yum repositories, click on Syst em → R ef resh p ackag e
list s to make sure your package list is up-to-date.

8.2.2. Finding Packages wit h Filt ers
Once the software sources have been updated, it is often beneficial to apply some filters so that
PackageKit retrieves the results of our Fin d queries faster. This is especially helpful when performing
many package searches. Four of the filters in the Filt ers drop-down menu are used to split results by
matching or not matching a single criterion. By default when PackageKit starts, these filters are all
unapplied (N o f ilt er), but once you do filter by one of them, that filter remains set until you either
change it or close PackageKit.
Because you are usually searching for available packages that are not installed on the system, click
Filt ers → In st alled and select the O n ly availab le radio button.

Fig u re 8.5. Filt erin g o u t alread y- in st alled p ackag es
Also, unless you require development files such as C header files, click Filt ers → D evelo p men t
and select the O n ly en d u ser f iles radio button. This filters out all of the - d evel
packages we are not interested in.

103

Deployment G uide

Fig u re 8.6 . Filt erin g o u t d evelo p men t p ackag es f ro m t h e list o f Fin d resu lt s
The two remaining filters with submenus are:
G rap h ical
Narrows the search to either applications which provide a GUI interface (O n ly g rap h ical)
or those that do not. This filter is useful when browsing for GUI applications that perform a
specific function.
Free
Search for packages which are considered to be free software. See the Fedora Licensing
List for details on approved licenses.
The remaining filters can be enabled by selecting the checkboxes next to them:
H id e su b p ackag es
Checking the H id e su b p ackag es checkbox filters out generally-uninteresting packages
that are typically only dependencies of other packages that we want. For example, checking
H id e su b p ackag es and searching for  would cause the following related
packages to be filtered out of the Fin d results (if it exists):
- d evel
- lib s
- lib s- d evel
- d eb u g in f o
O n ly n ewest p ackag es
Checking O n ly n ewest p ackag es filters out all older versions of the same package from
the list of results, which is generally what we want. Note that this filter is often combined with
the O n ly availab le filter to search for the latest available versions of new (not installed)
packages.

104

⁠Chapt er 8 . PackageKit

O n ly n at ive p ackag es
Checking the O n ly n at ive p ackag es box on a multilib system causes PackageKit to omit
listing results for packages compiled for the architecture that runs in compatibility mode. For
example, enabling this filter on a 64-bit system with an AMD 64 CPU would cause all
packages built for the 32-bit x86 CPU architecture not to be shown in the list of results,
even though those packages are able to run on an AMD 64 machine. Packages which are
architecture-agnostic (i.e. noarch packages such as cro n t ab s- 1.1032.1.el6 .n o arch .rp m) are never filtered out by checking O n ly n at ive p ackag es. This
filter has no affect on non-multilib systems, such as x86 machines.

8.2.3. Inst alling and Removing Packages (and Dependencies)
With the two filters selected, O n ly availab le and O n ly en d u ser f iles, search for the screen
window manager for the command line and highlight the package. You now have access to some
very useful information about it, including: a clickable link to the project homepage; the Yum package
group it is found in, if any; the license of the package; a pointer to the GNOME menu location from
where the application can be opened, if applicable; and the size of the package, which is relevant
when we download and install it.

Fig u re 8.7. Viewin g an d in st allin g a p ackag e wit h Packag eK it ' s Ad d /R emo ve So f t ware
win d o w
When the checkbox next to a package or group is checked, then that item is already installed on the
system. Checking an unchecked box causes it to be marked for installation, which only occurs when
the Ap p ly button is clicked. In this way, you can search for and select multiple packages or package
groups before performing the actual installation transactions. Additionally, you can remove installed
packages by unchecking the checked box, and the removal will occur along with any pending
installations when Ap p ly is pressed. D ependency resolution, which may add additional packages to
be installed or removed, is performed after pressing Ap p ly. PackageKit will then display a window
listing those additional packages to install or remove, and ask for confirmation to proceed.
Select screen and click the Ap p ly button. You will then be prompted for the superuser password;
enter it, and PackageKit will install screen . After finishing the installation, PackageKit sometimes
presents you with a list of your newly-installed applications and offers you the choice of running
them immediately. Alternatively, you will remember that finding a package and selecting it in the
Ad d /R emo ve So f t ware window shows you the Lo cat io n of where in the GNOME menus its
application shortcut is located, which is helpful when you want to run it.

105

Deployment G uide

Once it is installed, you can run screen , a screen manager that allows you to have multiple logins
on one terminal, by typing screen at a shell prompt.
screen is a very useful utility, but we decide that we do not need it and we want to uninstall it.
Remembering that we need to change the O n ly availab le filter we recently used to install it to O n ly
in st alled in Filt ers → In st alled , we search for screen again and uncheck it. The program did not
install any dependencies of its own; if it had, those would be automatically removed as well, as long
as they were not also dependencies of any other packages still installed on our system.

Removing a package when other packages depend on it
Although PackageKit automatically resolves dependencies during package installation and
removal, it is unable to remove a package without also removing packages which depend on
it. This type of operation can only be performed by R PM, is not advised, and can potentially
leave your system in a non-functioning state or cause applications to behave erratically
and/or crash.

Fig u re 8.8. R emo vin g a p ackag e wit h Packag eK it ' s Ad d /R emo ve So f t ware win d o w

8.2.4 . Inst alling and Removing Package Groups
PackageKit also has the ability to install Yum package groups, which it calls Packag e co llect io n s.
Clicking on Packag e co llect io n s in the top-left list of categories in the So f t ware U p d at es
window allows us to scroll through and find the package group we want to install. In this case, we
want to install Czech language support (the C z ech Su p p o rt group). Checking the box and clicking
ap p ly informs us how many additional packages must be installed in order to fulfill the dependencies
of the package group.

106

⁠Chapt er 8 . PackageKit

Fig u re 8.9 . In st allin g t h e C z ech Su p p o rt p ackag e g ro u p
Similarly, installed package groups can be uninstalled by selecting Packag e co llect io n s,
unchecking the appropriate checkbox, and applying.

8.2.5. Viewing t he T ransact ion Log
PackageKit maintains a log of the transactions that it performs. To view the log, from the
Ad d /R emo ve So f t ware window, click Syst em → So f t ware lo g , or run the g p k- lo g command at
the shell prompt.
The So f t ware Lo g Viewer shows the following information:
D at e — the date on which the transaction was performed.
Act io n — the action that was performed during the transaction, for example U p d at ed
p ackag es or In st alled p ackag es.
D et ails — the transaction type such as U p d at ed , In st alled , or R emo ved , followed by a list of
affected packages.
U sern ame — the name of the user who performed the action.
Ap p licat io n — the front end application that was used to perform the action, for example
U p d at e Syst em.
Typing the name of a package in the top text entry field filters the list of transactions to those which
affected that package.

107

Deployment G uide

Fig u re 8.10. Viewin g t h e lo g o f p ackag e man ag emen t t ran sact io n s wit h t h e So f t ware
Lo g Viewer

8.3. PackageKit Archit ect ure
Red Hat provides the PackageKit suite of applications for viewing, updating, installing and
uninstalling packages and package groups compatible with your system. Architecturally, PackageKit
consists of several graphical front ends that communicate with the p ackag ekit d daemon back end,
which communicates with a package manager-specific back end that utilizes Yum to perform the
actual transactions, such as installing and removing packages, etc.
Table 8.1, “ PackageKit GUI windows, menu locations, and shell prompt commands” shows the name
of the GUI window, how to start the window from the GNOME desktop or from the Ad d /R emo ve
So f t ware window, and the name of the command-line application that opens that window.
T ab le 8.1. Packag eK it G U I win d o ws, men u lo cat io n s, an d sh ell p ro mp t co mman d s
Win d o w T it le

Fu n ct io n

H o w t o O p en

Sh ell C o mman d

Add/Remove Software

Install, remove or view
package info

gpk-application

Software Update

Perform package
updates

Software Sources

Enable and disable
Yum repositories

Software Log Viewer

View the transaction
log

From the GNOME
panel: Syst em →
Ad min ist rat io n →
Ad d /R emo ve
So f t ware
From the GNOME
panel: Syst em →
Ad min ist rat io n →
So f t ware U p d at e
From Ad d /R emo ve
So f t ware: Syst em →
So f t ware So u rces
From Ad d /R emo ve
So f t ware: Syst em →
So f t ware Lo g

Software Update
Preferences

Set PackageKit
preferences

108

gpk-update-viewer

gpk-repo

gpk-log

gpk-prefs

⁠Chapt er 8 . PackageKit

Win d o w T it le

Fu n ct io n

(Notification Area Alert) Alerts you when
updates are available

H o w t o O p en

Sh ell C o mman d

From the GNOME
panel: Syst em →
Pref eren ces →
St art u p
Ap p licat io n s, the
St art u p Pro g rams
tab

gpk-update-icon

The p ackag ekit d daemon runs outside the user session and communicates with the various
graphical front ends. The p ackag ekit d daemon ⁠ [2] communicates via the D B u s system message
bus with another back end, which utilizes Yum's Python API to perform queries and make changes to
the system. On Linux systems other than Red Hat Enterprise Linux and Fedora, p ackag ekit d can
communicate with other back ends that are able to utilize the native package manager for that
system. This modular architecture provides the abstraction necessary for the graphical interfaces to
work with many different package managers to perform essentially the same types of package
management tasks. Learning how to use the PackageKit front ends means that you can use the same
familiar graphical interface across many different Linux distributions, even when they utilize a native
package manager other than Yum.
In addition, PackageKit's separation of concerns provides reliability in that a crash of one of the GUI
windows—or even the user's X Window session—will not affect any package management tasks
being supervised by the p ackag ekit d daemon, which runs outside of the user session.
All of the front end graphical applications discussed in this chapter are provided by the gnomepackagekit package instead of by PackageKit and its dependencies.
Finally, PackageKit also comes with a console-based front end called p kco n .

8.4 . Addit ional Resources
For more information about PackageKit, see the resources listed below.

Inst alle d Do cum e nt at io n
g p k- ap p licat io n ( 1) — The manual page containing information about the g p k- ap p licat io n
command.
g p k- b acken d - st at u s( 1) — The manual page containing information about the g p kb acken d - st at u s command.
g p k- in st all- lo cal- f ile( 1) — The manual page containing information about the g p k- in st alllo cal- f ile command.
g p k- in st all- mime- t yp e( 1) — The manual page containing information about the g p k- in st allmime- t yp e command.
g p k- in st all- p ackag e- n ame( 1) — The manual page containing information about the q p kin st all- p ackag e- n ame command.
g p k- in st all- p ackag e- n ame( 1) — The manual page containing information about the g p kin st all- p ackag e- n ame command.
g p k- p ref s( 1) — The manual page containing information about the g p k- p ref s command.
g p k- rep o ( 1) — The manual page containing information about the g p k- rep o command.

109

Deployment G uide

g p k- u p d at e- ico n ( 1) — The manual page containing information about the g p k- u p d at e- ico n
command.
g p k- u p d at e- viewer( 1) — The manual page containing information about the g p k- u p d at eviewer command.
p kco n ( 1) and p kmo n ( 1) — The manual pages containing information about the PackageKit
console client.
p kg en p ack( 1) — The manual page containing information about the PackageKit Pack
Generator.

Online Do cum e nt at io n
PackageKit home page — The PackageKit home page listing detailed information about the
PackageKit software suite.
PackageKit FAQ — An informative list of Frequently Asked Questions for the PackageKit software
suite.

Se e Also
Chapter 7, Yum documents Yum, the Red Hat package manager.

[2] Sys tem d aemo ns are typ ic ally lo ng -running p ro c es s es that p ro vid e s ervic es to the us er o r to o ther
p ro g rams , and whic h are s tarted , o ften at b o o t time, b y s p ec ial initializ atio n s c rip ts (o ften s ho rtened to
init scripts). Daemo ns res p o nd to the service c o mmand and c an b e turned o n o r o ff p ermanently b y
us ing the ch kco n f ig o n o r ch kco n f ig o f f c o mmand s . They c an typ ic ally b e rec o g niz ed b y a “ d”
ap p end ed to their name, s uc h as the p ackag ekit d d aemo n. See Chap ter 11, Services and Daemons fo r
info rmatio n ab o ut s ys tem s ervic es .

110

⁠P art IV. Net working

⁠Part IV. Networking
This part describes how to configure the network on Red Hat Enterprise Linux.

111

Deployment G uide

Chapter 9. NetworkManager
N et wo rkMan ag er is a dynamic network control and configuration system that attempts to keep
network devices and connections up and active when they are available. N et wo rkMan ag er
consists of a core daemon, a GNOME Notification Area applet that provides network status
information, and graphical configuration tools that can create, edit and remove connections and
interfaces. N et wo rkMan ag er can be used to configure the following types of connections: Ethernet,
wireless, mobile broadband (such as cellular 3G), and D SL and PPPo E (Point-to-Point over
Ethernet). In addition, N et wo rkMan ag er allows for the configuration of network aliases, static
routes, D NS information and VPN connections, as well as many connection-specific parameters.
Finally, N et wo rkMan ag er provides a rich API via D -Bus which allows applications to query and
control network configuration and state.
Previous versions of Red Hat Enterprise Linux included the N et wo rk Ad min ist rat io n T o o l, which
was commonly known as syst em- co n f ig - n et wo rk after its command-line invocation. In Red Hat
Enterprise Linux 6, N et wo rkMan ag er replaces the former N et wo rk Ad min ist rat io n T o o l while
providing enhanced functionality, such as user-specific and mobile broadband configuration. It is
also possible to configure the network in Red Hat Enterprise Linux 6 by editing interface
configuration files; see Chapter 10, Network Interfaces for more information.
N et wo rkMan ag er may be installed by default on your version of Red Hat Enterprise Linux. To
ensure that it is, run the following command as ro o t :
~]# yu m in st all N et wo rkMan ag er

9.1. T he Net workManager Daemon
The N et wo rkMan ag er daemon runs with ro o t privileges and is usually configured to start up at
boot time. You can determine whether the N et wo rkMan ag er daemon is running by entering this
command as ro o t :
~]# service N et wo rkMan ag er st at u s
NetworkManager (pid 1527) is running...
The service command will report N et wo rkMan ag er is st o p p ed if the N et wo rkMan ag er service
is not running. To start it for the current session:
~]# service N et wo rkMan ag er st art
Run the ch kco n f ig command to ensure that N et wo rkMan ag er starts up every time the system
boots:
~]# ch kco n f ig N et wo rkMan ag er o n
For more information on starting, stopping and managing services and runlevels, see Chapter 11,
Services and Daemons.

9.2. Int eract ing wit h Net workManager

112

⁠Chapt er 9 . Net workManager

Users do not interact with the N et wo rkMan ag er system service directly. Instead, you can perform
network configuration tasks via N et wo rkMan ag er's Notification Area applet. The applet has
multiple states that serve as visual indicators for the type of connection you are currently using.
Hover the pointer over the applet icon for tooltip information on the current connection state.

Fig u re 9 .1. N et wo rkMan ag er ap p let st at es
If you do not see the N et wo rkMan ag er applet in the GNOME panel, and assuming that the
N et wo rkMan ag er package is installed on your system, you can start the applet by running the
following command as a normal user (not ro o t ):
~]$ n m- ap p let &
After running this command, the applet appears in your Notification Area. You can ensure that the
applet runs each time you log in by clicking Syst em → Pref eren ces → St art u p Ap p licat io n s to
open the St art u p Ap p licat io n s Pref eren ces window. Then, select the St art u p Pro g rams tab
and check the box next to N et wo rkMan ag er.

9.2.1. Connect ing t o a Net work
When you left-click on the applet icon, you are presented with:
a list of categorized networks you are currently connected to (such as Wired and Wireless);
a list of all Availab le N et wo rks that N et wo rkMan ag er has detected;
options for connecting to any configured Virtual Private Networks (VPNs); and,
options for connecting to hidden or new wireless networks.
If you are connected to a network, its name is presented in bold typeface under its network type, such
as Wired or Wireless. When many networks are available, such as wireless access points, the
Mo re n et wo rks expandable menu entry appears.

113

Deployment G uide

Fig u re 9 .2. T h e N et wo rkMan ag er ap p let ' s lef t - click men u , sh o win g all availab le an d
co n n ect ed - t o n et wo rks

9.2.2. Configuring New and Edit ing Exist ing Connect ions
Next, right-click on the N et wo rkMan ag er applet to open its context menu, which is the main point of
entry for interacting with N et wo rkMan ag er to configure connections.

Fig u re 9 .3. T h e N et wo rkMan ag er ap p let ' s co n t ext men u
Ensure that the En ab le N et wo rkin g box is checked. If the system has detected a wireless card, then
you will also see an En ab le Wireless menu option. Check the En ab le Wireless checkbox as well.
N et wo rkMan ag er notifies you of network connection status changes if you check the En ab le
N o t if icat io n s box. Clicking the C o n n ect io n In f o rmat io n entry presents an informative
C o n n ect io n In f o rmat io n window that lists the connection type and interface, your IP address and
routing details, and so on.

114

⁠Chapt er 9 . Net workManager

Finally, clicking on Ed it C o n n ect io n s opens the N et wo rk C o n n ect io n s window, from where you
can perform most of your network configuration tasks. Note that this window can also be opened by
running, as a normal user:
~]$ n m- co n n ect io n - ed it o r &

Fig u re 9 .4 . C o n f ig u re n et wo rks u sin g t h e N et wo rk C o n n ect io n s win d o w
There is an arrow head symbol to the left which can be clicked to hide and reveal entries as needed.
To create a new connection, click the Ad d button to view the selection list, select the connection type
and click the C reat e button. Alternatively, to edit an existing connection select the interface name
from the list and click the Ed it button.
Then, to configure:
wired Ethernet connections, proceed to Section 9.3.1, “ Establishing a Wired (Ethernet)
Connection” ;
wireless connections, proceed to Section 9.3.2, “ Establishing a Wireless Connection” ; or,
mobile broadband connections, proceed to Section 9.3.3, “ Establishing a Mobile Broadband
Connection” ; or,
VPN connections, proceed to Section 9.3.4, “ Establishing a VPN Connection” .

9.2.3. Connect ing t o a Net work Aut omat ically
For any connection type you add or configure, you can choose whether you want
N et wo rkMan ag er to try to connect to that network automatically when it is available.
Pro ced u re 9 .1. C o n f ig u rin g N et wo rkMan ag er t o C o n n ect t o a N et wo rk Au t o mat ically
Wh en D et ect ed
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the arrow head if necessary to reveal the list of connections.
3. Select the specific connection that you want to configure and click Ed it .

115

Deployment G uide

4. Check C o n n ect au t o mat ically to cause N et wo rkMan ag er to auto-connect to the
connection whenever N et wo rkMan ag er detects that it is available. Uncheck the checkbox if
you do not want N et wo rkMan ag er to connect automatically. If the box is unchecked, you
will have to select that connection manually in the N et wo rkMan ag er applet's left-click menu
to cause it to connect.

9.2.4 . User and Syst em Connect ions
N et wo rkMan ag er connections are always either user connections or system connections. D epending
on the system-specific policy that the administrator has configured, users may need ro o t privileges
to create and modify system connections. N et wo rkMan ag er's default policy enables users to create
and modify user connections, but requires them to have ro o t privileges to add, modify or delete
system connections.
User connections are so-called because they are specific to the user who creates them. In contrast to
system connections, whose configurations are stored under the /et c/sysco n f ig /n et wo rk- scrip t s/
directory (mainly in if cf g -  interface configuration files), user connection settings are
stored in the GConf configuration database and the GNOME keyring, and are only available during
login sessions for the user who created them. Thus, logging out of the desktop session causes userspecific connections to become unavailable.

Increase security by making VPN connections user-specific
Because N et wo rkMan ag er uses the GConf and GNOME keyring applications to store user
connection settings, and because these settings are specific to your desktop session, it is
highly recommended to configure your personal VPN connections as user connections. If you
do so, other Non-ro o t users on the system cannot view or access these connections in any
way.
System connections, on the other hand, become available at boot time and can be used by other
users on the system without first logging in to a desktop session.
N et wo rkMan ag er can quickly and conveniently convert user to system connections and vice versa.
Converting a user connection to a system connection causes N et wo rkMan ag er to create the
relevant interface configuration files under the /et c/sysco n f ig /n et wo rk- scrip t s/ directory, and to
delete the GConf settings from the user's session. Conversely, converting a system to a user-specific
connection causes N et wo rkMan ag er to remove the system-wide configuration files and create the
corresponding GConf/GNOME keyring settings.

Fig u re 9 .5. T h e Availab le t o all u sers ch eckb o x co n t ro ls wh et h er co n n ect io n s are u sersp ecif ic o r syst em- wid e
Pro ced u re 9 .2. C h an g in g a C o n n ect io n t o b e U ser- Sp ecif ic in st ead o f Syst em- Wid e, o r
Vice- Versa

116

⁠Chapt er 9 . Net workManager

Root privileges may be required
D epending on the system's policy, you may need ro o t privileges on the system in order to
change whether a connection is user-specific or system-wide.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. If needed, select the arrow head (on the left hand side) to hide and reveal the types of
available network connections.
3. Select the specific connection that you want to configure and click Ed it .
4. Check the Availab le t o all u sers checkbox to ask N et wo rkMan ag er to make the
connection a system-wide connection. D epending on system policy, you may then be
prompted for the ro o t password by the Po licyK it application. If so, enter the ro o t password
to finalize the change.
Conversely, uncheck the Availab le t o all u sers checkbox to make the connection userspecific.

9.3. Est ablishing Connect ions
9.3.1. Est ablishing a Wired (Et hernet ) Connect ion
To establish a wired network connection, Right-click on the N et wo rkMan ag er applet to open its
context menu, ensure that the En ab le N et wo rkin g box is checked, then click on Ed it
C o n n ect io n s. This opens the N et wo rk C o n n ect io n s window. Note that this window can also be
opened by running, as a normal user:
~]$ n m- co n n ect io n - ed it o r &
You can click on the arrow head to reveal and hide the list of connections as needed.

Fig u re 9 .6 . T h e N et wo rk C o n n ect io n s win d o w sh o win g t h e n ewly creat ed Syst em et h 0
co n n ect io n

117

Deployment G uide

The system startup scripts create and configure a single wired connection called Syst em et h 0 by
default on all systems. Although you can edit Syst em et h 0, creating a new wired connection for
your custom settings is recommended. You can create a new wired connection by clicking the Ad d
button, selecting the Wired entry from the list that appears and then clicking the C reat e button.

Fig u re 9 .7. Select in g a n ew co n n ect io n t yp e f ro m t h e "C h o o se a C o n n ect io n T yp e" list

T he dialog for adding and editing connections is the same
When you add a new connection by clicking the Ad d button, a list of connection types
appears. Once you have made a selection and clicked on the C reat e button,
N et wo rkMan ag er creates a new configuration file for that connection and then opens the
same dialog that is used for editing an existing connection. There is no difference between
these dialogs. In effect, you are always editing a connection; the difference only lies in whether
that connection previously existed or was just created by N et wo rkMan ag er when you
clicked C reat e.

118

⁠Chapt er 9 . Net workManager

Fig u re 9 .8. Ed it in g t h e n ewly creat ed Wired co n n ect io n Syst em et h 0

Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name will be
used to list this connection in the Wired section of the N et wo rk C o n n ect io n s window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 9.2.3, “ Connecting to a Network Automatically”
for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on the
system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User and System
Connections” for details.

Co nfiguring t he Wire d T ab
The final three configurable settings are located within the Wired tab itself: the first is a text-entry field
where you can specify a MAC (Media Access Control) address, and the second allows you to specify
a cloned MAC address, and third allows you to specify the MTU (Maximum Transmission Unit) value.
Normally, you can leave the MAC ad d ress field blank and the MT U set to au t o mat ic. These
defaults will suffice unless you are associating a wired connection with a second or specific NIC, or
performing advanced networking. In such cases, see the following descriptions:

119

Deployment G uide

MAC Ad d ress
Network hardware such as a Network Interface Card (NIC) has a unique MAC address
(Media Access Control; also known as a hardware address) that identifies it to the system.
Running the ip ad d r command will show the MAC address associated with each interface.
For example, in the following ip ad d r output, the MAC address for the et h 0 interface (which
is 52:54 :00:26 :9 e:f 1) immediately follows the lin k/et h er keyword:
~]# ip ad d r
1: lo:  mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast state
UNKNOWN qlen 1000
link/ether 52:54:00:26:9e:f1 brd ff:ff:ff:ff:ff:ff
inet 192.168.122.251/24 brd 192.168.122.255 scope global eth0
inet6 fe80::5054:ff:fe26:9ef1/64 scope link
valid_lft forever preferred_lft forever
A single system can have one or more NICs installed on it. The MAC ad d ress field
therefore allows you to associate a specific NIC with a specific connection (or connections).
As mentioned, you can determine the MAC address using the ip ad d r command, and then
copy and paste that value into the MAC ad d ress text-entry field.
The cloned MAC address field is mostly for use in such situations were a network service
has been restricted to a specific MAC address and you need to emulate that MAC address.
MT U
The MTU (Maximum Transmission Unit) value represents the size in bytes of the largest
packet that the connection will use to transmit. This value defaults to 1500 when using
IPv4, or a variable number 1280 or higher for IPv6, and does not generally need to be
specified or changed.

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your wired connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can connect to your new or customized connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
port-based Network Access Control (PNAC), click the 802.1X Secu rit y tab and proceed to
Section 9.3.9.1, “ Configuring 802.1X Security” ;
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IPv6 Set t in g s tab and proceed to Section 9.3.9.5,
“ Configuring IPv6 Settings” .

120

⁠Chapt er 9 . Net workManager

9.3.2. Est ablishing a Wireless Connect ion
This section explains how to use N et wo rkMan ag er to configure a wireless (also known as Wi-Fi or
802.11a/b/g/n) connection to an Access Point.
To configure a mobile broadband (such as 3G) connection, see Section 9.3.3, “ Establishing a
Mobile Broadband Connection” .

Quickly Co nne ct ing t o an Available Acce ss Po int
The easiest way to connect to an available access point is to left-click on the N et wo rkMan ag er
applet, locate the Service Set Identifier (SSID ) of the access point in the list of Availab le networks,
and click on it. If the access point is secured, a dialog prompts you for authentication.

Fig u re 9 .9 . Au t h en t icat in g t o a wireless access p o in t
N et wo rkMan ag er tries to auto-detect the type of security used by the access point. If there are
multiple possibilities, N et wo rkMan ag er guesses the security type and presents it in the Wireless
secu rit y dropdown menu. To see if there are multiple choices, click the Wireless secu rit y
dropdown menu and select the type of security the access point is using. If you are unsure, try
connecting to each type in turn. Finally, enter the key or passphrase in the Passwo rd field. Certain
password types, such as a 40-bit WEP or 128-bit WPA key, are invalid unless they are of a requisite
length. The C o n n ect button will remain inactive until you enter a key of the length required for the
selected security type. To learn more about wireless security, see Section 9.3.9.2, “ Configuring
Wireless Security” .

121

Deployment G uide

Prevent Roaming On T he Same Access Point
In the case of WPA and WPA2 (Personal and Enterprise), an option to select between Auto,
WPA and WPA2 has been added. This option is intended for use with an access point that is
offering both WPA and WPA2. Select one of the protocols if you would like to prevent roaming
between the two protocols. Roaming between WPA and WPA2 on the same access point can
cause loss of service.
If N et wo rkMan ag er connects to the access point successfully, its applet icon will change into a
graphical indicator of the wireless connection's signal strength.

Fig u re 9 .10. Ap p let ico n in d icat in g a wireless co n n ect io n sig n al st ren g t h o f 75%
You can also edit the settings for one of these auto-created access point connections just as if you
had added it yourself. The Wireless tab of the N et wo rk C o n n ect io n s window lists all of the
connections you have ever tried to connect to: N et wo rkMan ag er names each of them Au t o ,
where SSID is the Service Set identifier of the access point.

Fig u re 9 .11. An examp le o f access p o in t s t h at h ave p revio u sly b een co n n ect ed t o

Co nne ct ing t o a Hidde n Wire le ss Ne t wo rk

122

⁠Chapt er 9 . Net workManager

All access points have a Service Set Identifier (SSID ) to identify them. However, an access point may
be configured not to broadcast its SSID , in which case it is hidden, and will not show up in
N et wo rkMan ag er's list of Availab le networks. You can still connect to a wireless access point that
is hiding its SSID as long as you know its SSID , authentication method, and secrets.
To connect to a hidden wireless network, left-click N et wo rkMan ag er's applet icon and select
C o n n ect t o H id d en Wireless N et wo rk to cause a dialog to appear. If you have connected to the
hidden network before, use the C o n n ect io n dropdown to select it, and click C o n n ect . If you have
not, leave the C o n n ect io n dropdown as N ew, enter the SSID of the hidden network, select its
Wireless secu rit y method, enter the correct authentication secrets, and click C o n n ect .
For more information on wireless security settings, see Section 9.3.9.2, “ Configuring Wireless
Security” .

Edit ing a Co nne ct io n, o r Cre at ing a Co m ple t e ly Ne w One
You can edit an existing connection that you have tried or succeeded in connecting to in the past by
opening the Wireless tab of the N et wo rk C o n n ect io n s, selecting the connection by name (words
which follow Au t o refer to the SSID of an access point), and clicking Ed it .
You can create a new connection by opening the N et wo rk C o n n ect io n s window, clicking the Ad d
button, selecting Wireless, and clicking the C reat e button.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button.
3. Select the Wireless entry from the list.
4. Click the C reat e button.

123

Deployment G uide

Fig u re 9 .12. Ed it in g t h e n ewly creat ed Wireless co n n ect io n 1

Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name will be
used to list this connection in the Wireless section of the N et wo rk C o n n ect io n s window. By
default, wireless connections are named the same as the SSID of the wireless access point. You
can rename the wireless connection without affecting its ability to connect, but it is recommended
to retain the SSID name.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 9.2.3, “ Connecting to a Network Automatically”
for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on the
system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User and System
Connections” for details.

124

⁠Chapt er 9 . Net workManager

Co nfiguring t he Wire le ss T ab
SSID
All access points have a Service Set identifier to identify them. However, an access point may
be configured not to broadcast its SSID , in which case it is hidden, and will not show up in
N et wo rkMan ag er's list of Availab le networks. You can still connect to a wireless access
point that is hiding its SSID as long as you know its SSID (and authentication secrets).
For information on connecting to a hidden wireless network, see Section 9.3.2, “ Connecting
to a Hidden Wireless Network” .
Mo d e
In f rast ru ct u re — Set Mo d e to In f rast ru ct u re if you are connecting to a dedicated
wireless access point or one built into a network device such as a router or a switch.
Ad - h o c — Set Mo d e to Ad - h o c if you are creating a peer-to-peer network for two or more
mobile devices to communicate directly with each other. If you use Ad - h o c mode, referred
to as Independent Basic Service Set (IBSS) in the 802.11 standard, you must ensure that the
same SSID is set for all participating wireless devices, and that they are all communicating
over the same channel.
B SSID
The Basic Service Set Identifier (BSSID ) is the MAC address of the specific wireless access
point you are connecting to when in In f rast ru ct u re mode. This field is blank by default,
and you are able to connect to a wireless access point by SSID without having to specify
its B SSID . If the BSSID is specified, it will force the system to associate to a specific access
point only.
For ad-hoc networks, the B SSID is generated randomly by the mac80211 subsystem when
the ad-hoc network is created. It is not displayed by N et wo rkMan ag er
MAC ad d ress
Like an Ethernet Network Interface Card (NIC), a wireless adapter has a unique MAC
address (Media Access Control; also known as a hardware address) that identifies it to the
system. Running the ip ad d r command will show the MAC address associated with each
interface. For example, in the following ip ad d r output, the MAC address for the wlan 0
interface (which is 00:1c:b f :02:f 8:70) immediately follows the lin k/et h er keyword:
~]# ip ad d r
1: lo:  mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast state
UNKNOWN qlen 1000
link/ether 52:54:00:26:9e:f1 brd ff:ff:ff:ff:ff:ff
inet 192.168.122.251/24 brd 192.168.122.255 scope global eth0
inet6 fe80::5054:ff:fe26:9ef1/64 scope link
valid_lft forever preferred_lft forever
3: wlan0:  mtu 1500 qdisc mq state UP
qlen 1000

125

Deployment G uide

link/ether 00:1c:bf:02:f8:70 brd ff:ff:ff:ff:ff:ff
inet 10.200.130.67/24 brd 10.200.130.255 scope global wlan0
inet6 fe80::21c:bfff:fe02:f870/64 scope link
valid_lft forever preferred_lft forever
A single system could have one or more wireless network adapters connected to it. The
MAC ad d ress field therefore allows you to associate a specific wireless adapter with a
specific connection (or connections). As mentioned, you can determine the MAC address
using the ip ad d r command, and then copy and paste that value into the MAC ad d ress
text-entry field.
MT U
The MTU (Maximum Transmission Unit) value represents the size in bytes of the largest
packet that the connection will use to transmit. If set to a non-zero number, only packets of
the specified size or smaller will be transmitted. Larger packets are broken up into multiple
Ethernet frames. It is recommended to leave this setting on au t o mat ic.

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing the wireless connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can successfully connect to your the modified connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for details
on selecting and connecting to a network.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
security authentication for the wireless connection, click the Wireless Secu rit y tab and proceed
to Section 9.3.9.2, “ Configuring Wireless Security” ;
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IPv6 Set t in g s tab and proceed to Section 9.3.9.5,
“ Configuring IPv6 Settings” .

9.3.3. Est ablishing a Mobile Broadband Connect ion
You can use N et wo rkMan ag er's mobile broadband connection abilities to connect to the following
2G and 3G services:
2G — GPRS (General Packet Radio Service) or EDGE (Enhanced Data Rates for GSM Evolution)
3G — UMTS (Universal Mobile Telecommunications System) or HSPA (High Speed Packet Access)
Your computer must have a mobile broadband device (modem), which the system has discovered
and recognized, in order to create the connection. Such a device may be built into your computer (as
is the case on many notebooks and netbooks), or may be provided separately as internal or external
hardware. Examples include PC card, USB Modem or D ongle, mobile or cellular telephone capable
of acting as a modem.
Pro ced u re 9 .3. Ad d in g a N ew Mo b ile B ro ad b an d C o n n ect io n

126

⁠Chapt er 9 . Net workManager

You can configure a mobile broadband connection by opening the N et wo rk C o n n ect io n s
window, clicking Ad d , and selecting Mo b ile B ro ad b an d from the list.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button to open the selection list. Select Mo b ile B ro ad b an d and then click
C reat e. The Set u p a Mo b ile B ro ad b an d C o n n ect io n assistant appears.
3. Under C reat e a co n n ect io n f o r t h is mo b ile b ro ad b an d d evice, choose the 2G- or 3Gcapable device you want to use with the connection. If the dropdown menu is inactive, this
indicates that the system was unable to detect a device capable of mobile broadband. In this
case, click C an cel, ensure that you do have a mobile broadband-capable device attached
and recognized by the computer and then retry this procedure. Click the Fo rward button.
4. Select the country where your service provider is located from the list and click the Fo rward
button.
5. Select your provider from the list or enter it manually. Click the Fo rward button.
6. Select your payment plan from the dropdown menu and confirm the Access Point Name (APN)
is correct. Click the Fo rward button.
7. Review and confirm the settings and then click the Ap p ly button.
8. Edit the mobile broadband-specific settings by referring to the Configuring the Mobile Broadband
Tab description below .
Pro ced u re 9 .4 . Ed it in g an Exist in g Mo b ile B ro ad b an d C o n n ect io n
Follow these steps to edit an existing mobile broadband connection.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you wish to edit and click the Ed it button.
3. Select the Mo b ile B ro ad b an d tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name
will be used to list this connection in the Mo b ile B ro ad b an d section of the N et wo rk
C o n n ect io n s window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to autoconnect to this connection when it is available. See Section 9.2.3, “ Connecting to a
Network Automatically” for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on
the system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User
and System Connections” for details.
5. Edit the mobile broadband-specific settings by referring to the Configuring the Mobile Broadband
Tab description below .

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns

127

Deployment G uide

Once you have finished editing your mobile broadband connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can connect to your new or customized connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
Point-to-point settings for the connection, click the PPP Set t in g s tab and proceed to
Section 9.3.9.3, “ Configuring PPP (Point-to-Point) Settings” ;
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IPv6 Set t in g s tab and proceed to Section 9.3.9.5,
“ Configuring IPv6 Settings” .

Co nfiguring t he Mo bile Bro adband T ab
If you have already added a new mobile broadband connection using the assistant (see
Procedure 9.3, “ Adding a New Mobile Broadband Connection” for instructions), you can edit the
Mo b ile B ro ad b an d tab to disable roaming if home network is not available, assign a network ID , or
instruct N et wo rkMan ag er to prefer a certain technology (such as 3G or 2G) when using the
connection.
N u mb er
The number that is dialed to establish a PPP connection with the GSM-based mobile
broadband network. This field may be automatically populated during the initial installation
of the broadband device. You can usually leave this field blank and enter the APN instead.
U sern ame
Enter the user name used to authenticate with the network. Some providers do not provide a
user name, or accept any user name when connecting to the network.
Passwo rd
Enter the password used to authenticate with the network. Some providers do not provide a
password, or accept any password.
APN
Enter the Access Point Name (APN) used to establish a connection with the GSM-based
network. Entering the correct APN for a connection is important because it often determines:
how the user is billed for their network usage; and/or
whether the user has access to the Internet, an intranet, or a subnetwork.
N et wo rk ID
Entering a N et wo rk ID causes N et wo rkMan ag er to force the device to register only to a
specific network. This can be used to ensure the connection does not roam when it is not
possible to control roaming directly.
T yp e

128

⁠Chapt er 9 . Net workManager

An y — The default value of An y leaves the modem to select the fastest network.
3G ( U MT S/H SPA) — Force the connection to use only 3G network technologies.
2G ( G PR S/ED G E) — Force the connection to use only 2G network technologies.
Pref er 3G ( U MT S/H SPA) — First attempt to connect using a 3G technology such as
HSPA or UMTS, and fall back to GPRS or ED GE only upon failure.
Pref er 2G ( G PR S/ED G E) — First attempt to connect using a 2G technology such as
GPRS or ED GE, and fall back to HSPA or UMTS only upon failure.
Allo w ro amin g if h o me n et wo rk is n o t availab le
Uncheck this box if you want N et wo rkMan ag er to terminate the connection rather than
transition from the home network to a roaming one, thereby avoiding possible roaming
charges. If the box is checked, N et wo rkMan ag er will attempt to maintain a good
connection by transitioning from the home network to a roaming one, and vice versa.
PIN
If your device's SIM (Subscriber Identity Module) is locked with a PIN (Personal Identification
Number), enter the PIN so that N et wo rkMan ag er can unlock the device.
N et wo rkMan ag er must unlock the SIM if a PIN is required in order to use the device for
any purpose.

9.3.4 . Est ablishing a VPN Connect ion
Establishing an encrypted Virtual Private Network (VPN) enables you to communicate securely
between your Local Area Network (LAN), and another, remote LAN. After successfully establishing a
VPN connection, a VPN router or gateway performs the following actions upon the packets you
transmit:
1. it adds an Authentication Header for routing and authentication purposes;
2. it encrypts the packet data; and,
3. it encloses the data with an Encapsulating Security Payload (ESP), which constitutes the
decryption and handling instructions.
The receiving VPN router strips the header information, decrypts the data, and routes it to its intended
destination (either a workstation or other node on a network). Using a network-to-network
connection, the receiving node on the local network receives the packets already decrypted and
ready for processing. The encryption/decryption process in a network-to-network VPN connection is
therefore transparent to clients.
Because they employ several layers of authentication and encryption, VPNs are a secure and
effective means of connecting multiple remote nodes to act as a unified intranet.
Pro ced u re 9 .5. Ad d in g a N ew VPN C o n n ect io n
1. You can configure a new VPN connection by opening the N et wo rk C o n n ect io n s window,
clicking the Ad d button and selecting a type of VPN from the VPN section of the new
connection list.
2. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
3. Click the Ad d button.

129

Deployment G uide

4. The C h o o se a C o n n ect io n T yp e list appears.

5.

A VPN plug-in is required
The appropriate N et wo rkMan ag er VPN plug-in for the VPN type you want to
configure must be installed (see Section 7.2.4, “ Installing Packages” for more
information on how to install new packages in Red Hat Enterprise Linux 6).
The VPN section in the C h o o se a C o n n ect io n T yp e list will n o t appear if you do
not have a suitable plug-in installed.

6. ⁠
Select the VPN protocol for the gateway you are connecting to from the C h o o se a
C o n n ect io n T yp e list. The VPN protocols available for selection in the list correspond to
the N et wo rkMan ag er VPN plug-ins installed. For example, if the N et wo rkMan ag er VPN
plug-in for openswan, NetworkManager-openswan, is installed then the IPsec based VPN will
be selectable from the C h o o se a C o n n ect io n T yp e list.
After selecting the correct one, press the C reat e button.
7. The Ed it in g VPN C o n n ect io n 1 window then appears. This window presents settings
customized for the type of VPN connection you selected in Step 6.
Pro ced u re 9 .6 . Ed it in g an Exist in g VPN C o n n ect io n
You can configure an existing VPN connection by opening the N et wo rk C o n n ect io n s window and
selecting the name of the connection from the list. Then click the Ed it button.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you wish to edit and click the Ed it button.

130

⁠Chapt er 9 . Net workManager

Fig u re 9 .13. Ed it in g t h e n ewly creat ed IPsec VPN co n n ect io n 1

Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name will be
used to list this connection in the VPN section of the N et wo rk C o n n ect io n s window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 9.2.3, “ Connecting to a Network Automatically”
for more information.

131

Deployment G uide

Availab le t o all u sers — Check this box to create a connection available to all users on the
system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User and System
Connections” for details.

Co nfiguring t he VPN T ab
G at eway
The name or IP address of the remote VPN gateway.
G ro u p n ame
The name of a VPN group configured on the remote gateway.
U ser p asswo rd
If required, enter the password used to authenticate with the VPN.
G ro u p p asswo rd
If required, enter the password used to authenticate with the VPN.
U ser n ame
If required, enter the user name used to authenticate with the VPN.
Ph ase1 Alg o rit h ms
If required, enter the algorithms to be used to authenticate and set up an encrypted
channel.
Ph ase2 Alg o rit h ms
If required, enter the algorithms to be used for the IPsec negotiations.
D o main
If required, enter the D omain Name.
N AT t raversal
C isco U D P ( d ef au lt ) — IPsec over UD P.
N AT - T — ESP encapsulation and IKE extensions are used to handle NAT Traversal.
D isab led — No special NAT measures required.
D isab le D ead Peer D et ect io n — D isable the sending of probes to the remote gateway or
endpoint.

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your new VPN connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can connect to your new or customized connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.

132

⁠Chapt er 9 . Net workManager

Then, to configure:
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” .

9.3.5. Est ablishing a DSL Connect ion
This section is intended for those installations which have a D SL card fitted within a host rather than
the external combined D SL modem router combinations typical of private consumer or SOHO
installations.
Pro ced u re 9 .7. Ad d in g a N ew D SL C o n n ect io n
You can configure a new D SL connection by opening the N et wo rk C o n n ect io n s window, clicking
the Ad d button and selecting D SL from the H ard ware section of the new connection list.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button.
3. The C h o o se a C o n n ect io n T yp e list appears.
4. Select D SL and press the C reat e button.
5. The Ed it in g D SL C o n n ect io n 1 window appears.
Pro ced u re 9 .8. Ed it in g an Exist in g D SL C o n n ect io n
You can configure an existing D SL connection by opening the N et wo rk C o n n ect io n s window and
selecting the name of the connection from the list. Then click the Ed it button.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you wish to edit and click the Ed it button.

Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name will be
used to list this connection in the D SL section of the N et wo rk C o n n ect io n s window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 9.2.3, “ Connecting to a Network Automatically”
for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on the
system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User and System
Connections” for details.

Co nfiguring t he DSL T ab
U sern ame
Enter the user name used to authenticate with the service provider.

133

Deployment G uide

Service
Leave blank unless otherwise directed.
Passwo rd
Enter the password supplied by the service provider.

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your D SL connection, click the Ap p ly button and N et wo rkMan ag er
will immediately save your customized configuration. Given a correct configuration, you can connect
to your new or customized connection by selecting it from the N et wo rkMan ag er Notification Area
applet. See Section 9.2.1, “ Connecting to a Network” for information on using your new or altered
connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
The MAC address and MTU settings, click the Wired tab and proceed to Section 9.3.1,
“ Configuring the Wired Tab” ;
Point-to-point settings for the connection, click the PPP Set t in g s tab and proceed to
Section 9.3.9.3, “ Configuring PPP (Point-to-Point) Settings” ;
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” .

9.3.6. Est ablishing a Bond Connect ion
You can use N et wo rkMan ag er to create a Bond from two or more Wired or Infiniband connections.
It is not necessary to create the connections to be bonded first. They can be configured as part of the
process to configure the bond. You must have the MAC addresses of the interfaces available in order
to complete the configuration process.

Note
N et wo rkMan ag er support for bonding must be enabled by means of the
N M_B O N D _VLAN _EN AB LED directive and then N et wo rkMan ag er must be restarted. See
Section 10.2.1, “ Ethernet Interfaces” for an explanation of N M_C O N T R O LLED and the
N M_B O N D _VLAN _EN AB LED directive. See Section 11.3.4, “ Restarting a Service” for an
explantion of restarting a service such as N et wo rkMan ag er from the command line.
Alternatively, for a graphical tool see Section 11.2.1, “ Using the Service Configuration Utility” .

Pro ced u re 9 .9 . Ad d in g a N ew B o n d C o n n ect io n
You can configure a Bond connection by opening the N et wo rk C o n n ect io n s window, clicking
Ad d , and selecting B o n d from the list.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button to open the selection list. Select B o n d and then click C reat e. The
Ed it in g B o n d co n n ect io n 1 window appears.

134

⁠Chapt er 9 . Net workManager

3. On the B o n d tab, click Ad d and select the type of interface you want to use with the bond
connection. Click the C reat e button. Note that the dialog to select the slave type only comes
up when you create the first slave; after that, it will automatically use that same type for all
further slaves.
4. The Ed it in g b o n d 0 slave 1 window appears. Fill in the MAC address of the first interface to
be bonded. The first slave's MAC address will be used as the MAC address for the bond
interface. If required, enter a clone MAC address to be used as the bond's MAC address. Click
the Ap p ly button.
5. The Au t h en t icat e window appears. Enter the ro o t password to continue. Click the
Au t h en t icat e button.
6. The name of the bonded slave appears in the B o n d ed C o n n ect io n s win d o w. Click the
Ad d button to add further slave connections.
7. Review and confirm the settings and then click the Ap p ly button.
8. Edit the bond-specific settings by referring to Section 9.3.6, “ Configuring the Bond Tab”
below.

135

Deployment G uide

Fig u re 9 .14 . Ed it in g t h e n ewly creat ed B o n d co n n ect io n 1
Pro ced u re 9 .10. Ed it in g an Exist in g B o n d C o n n ect io n
Follow these steps to edit an existing bond connection.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you want to edit and click the Ed it button.
3. Select the B o n d tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Three settings in the Ed it in g dialog are common to all connection types:

136

⁠Chapt er 9 . Net workManager

C o n n ect io n n ame — Enter a descriptive name for your network connection. This name
will be used to list this connection in the B o n d section of the N et wo rk C o n n ect io n s
window.
C o n n ect au t o mat ically — Select this box if you want N et wo rkMan ag er to autoconnect to this connection when it is available. See Section 9.2.3, “ Connecting to a
Network Automatically” for more information.
Availab le t o all u sers — Select this box to create a connection available to all users on
the system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User
and System Connections” for details.
5. Edit the bond-specific settings by referring to Section 9.3.6, “ Configuring the Bond Tab”
below.

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your bond connection, click the Ap p ly button to save your
customized configuration. Given a correct configuration, you can connect to your new or customized
connection by selecting it from the N et wo rkMan ag er Notification Area applet. See Section 9.2.1,
“ Connecting to a Network” for information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IPv6 Set t in g s tab and proceed to Section 9.3.9.5,
“ Configuring IPv6 Settings” .

Co nfiguring t he Bo nd T ab
If you have already added a new bond connection (see Procedure 9.9, “ Adding a New Bond
Connection” for instructions), you can edit the B o n d tab to set the load sharing mode and the type
of link monitoring to use to detect failures of a slave connection.
Mo d e
The mode that is used to share traffic over the slave connections which make up the bond.
The default is R o u n d - ro b in . Other load sharing modes, such as 802.3ad , can be
selected by means of the drop-down list.
Lin k Mo n it o rin g
The method of monitoring the slaves ability to carry network traffic.
The following modes of load sharing are selectable from the Mo d e drop-down list:
R o u n d - ro b in
Sets a round-robin policy for fault tolerance and load balancing. Transmissions are
received and sent out sequentially on each bonded slave interface beginning with the first
one available. This mode might not work behind a bridge with virtual machines without
additional switch configuration.
Act ive b acku p

137

Deployment G uide

Sets an active-backup policy for fault tolerance. Transmissions are received and sent out
via the first available bonded slave interface. Another bonded slave interface is only used if
the active bonded slave interface fails. Note that this is the only mode available for bonds of
InfiniBand devices.
XO R
Sets an XOR (exclusive-or) policy. Transmissions are based on the selected hash policy.
The default is to derive a hash by XOR of the source and destination MAC addresses
multiplied by the modulo of the number of slave interfaces. In this mode traffic destined for
specific peers will always be sent over the same interface. As the destination is determined
by the MAC addresses this method works best for traffic to peers on the same link or local
network. If traffic has to pass through a single router then this mode of traffic balancing will
be suboptimal.
B ro ad cast
Sets a broadcast policy for fault tolerance. All transmissions are sent on all slave
interfaces. This mode might not work behind a bridge with virtual machines without
additional switch configuration.
802.3ad
Sets an IEEE 802.3ad dynamic link aggregation policy. Creates aggregation groups that
share the same speed and duplex settings. Transmits and receives on all slaves in the
active aggregator. Requires a network switch that is 802.3ad compliant.
Ad ap t ive t ran smit lo ad b alan cin g
Sets an adaptive Transmit Load Balancing (TLB) policy for fault tolerance and load
balancing. The outgoing traffic is distributed according to the current load on each slave
interface. Incoming traffic is received by the current slave. If the receiving slave fails,
another slave takes over the MAC address of the failed slave. This mode is only suitable for
local addresses known to the kernel bonding module and therefore cannot be used behind
a bridge with virtual machines.
Ad ap t ive lo ad b alan cin g
Sets an Adaptive Load Balancing (ALB) policy for fault tolerance and load balancing.
Includes transmit and receive load balancing for IPv4 traffic. Receive load balancing is
achieved through AR P negotiation. This mode is only suitable for local addresses known
to the kernel bonding module and therefore cannot be used behind a bridge with virtual
machines.
The following types of link monitoring can be selected from the Lin k Mo n it o rin g drop-down list. It is
a good idea to test which channel bonding module parameters work best for your bonded interfaces.
MII ( Med ia In d ep en d en t In t erf ace)
The state of the carrier wave of the interface is monitored. This can be done by querying the
driver, by querying MII registers directly, or by using et h t o o l to query the device. Three
options are available:
Mo n it o rin g Freq u en cy
The time interval, in milliseconds, between querying the driver or MII registers.
Lin k u p d elay

138

⁠Chapt er 9 . Net workManager

The time in milliseconds to wait before attempting to use a link that has been
reported as up. This delay can be used if some gratuitous AR P requests are lost
in the period immediately following the link being reported as “ up” . This can
happen during switch initialization for example.
Lin k d o wn d elay
The time in milliseconds to wait before changing to another link when a previously
active link has been reported as “ down” . This delay can be used if an attached
switch takes a relatively long time to change to backup mode.
AR P
The address resolution protocol (AR P) is used to probe one or more peers to determine
how well the link-layer connections are working. It is dependent on the device driver
providing the transmit start time and the last receive time.
Two options are available:
Mo n it o rin g Freq u en cy
The time interval, in milliseconds, between sending AR P requests.
AR P t arg et s
A comma separated list of IP addresses to send AR P requests to.

9.3.7. Est ablishing a VLAN Connect ion
You can use N et wo rkMan ag er to create a VLAN using an existing interface. Currently, at time of
writing, you can only make VLANs on Ethernet devices.
Pro ced u re 9 .11. Ad d in g a N ew VLAN C o n n ect io n
You can configure a VLAN connection by opening the N et wo rk C o n n ect io n s window, clicking
Ad d , and selecting VLAN from the list.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button to open the selection list. Select VLAN and then click C reat e. The
Ed it in g VLAN C o n n ect io n 1 window appears.
3. On the VLAN tab, select the parent interface from the drop-down list you want to use for the
VLAN connection.
4. Enter the VLAN ID
5. Enter a VLAN interface name. This is the name of the VLAN interface that will be created. For
example, " eth0.1" or " vlan2" . (Normally this is either the parent interface name plus " ." and
the VLAN ID , or " vlan" plus the VLAN ID .)
6. Review and confirm the settings and then click the Ap p ly button.
7. Edit the VLAN-specific settings by referring to the Configuring the VLAN Tab description below .
Pro ced u re 9 .12. Ed it in g an Exist in g VLAN C o n n ect io n
Follow these steps to edit an existing VLAN connection.

139

Deployment G uide

1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you wish to edit and click the Ed it button.
3. Select the VLAN tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name
will be used to list this connection in the VLAN section of the N et wo rk C o n n ect io n s
window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to autoconnect to this connection when it is available. See Section 9.2.3, “ Connecting to a
Network Automatically” for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on
the system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User
and System Connections” for details.
5. Edit the VLAN-specific settings by referring to the Configuring the VLAN Tab description below .

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your VLAN connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can connect to your new or customized connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” .

Co nfiguring t he VLAN T ab
If you have already added a new VLAN connection (see Procedure 9.11, “ Adding a New VLAN
Connection” for instructions), you can edit the VLAN tab to set the parent interface and the VLAN ID .
Paren t In t erf ace
A previously configured interface can be selected in the drop-down list.
VLAN ID
The identification number to be used to tag the VLAN network traffic.
VLAN in t erf ace n ame
The name of the VLAN interface that will be created. For example, " eth0.1" or " vlan2" .
C lo n ed MAC ad d ress

14 0

⁠Chapt er 9 . Net workManager

Optionally sets an alternate MAC address to use for identifying the VLAN interface. This can
be used to change the source MAC address for packets sent on this VLAN.
MT U
Optionally sets a Maximum Transmission Unit (MTU) size to be used for packets to be sent
over the VLAN connection.

9.3.8. Est ablishing an IP-over-InfiniBand (IPoIB) Connect ion
You can use N et wo rkMan ag er to create an InfiniBand connection.
Pro ced u re 9 .13. Ad d in g a N ew In f in iB an d C o n n ect io n
You can configure an InfiniBand connection by opening the N et wo rk C o n n ect io n s window,
clicking Ad d , and selecting In f in iB an d from the list.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Click the Ad d button to open the selection list. Select In f in iB an d and then click C reat e. The
Ed it in g In f in iB an d C o n n ect io n 1 window appears.
3. On the In f in iB an d tab, select the transport mode from the drop-down list you want to use for
the InfiniBand connection.
4. Enter the InfiniBand MAC address.
5. Review and confirm the settings and then click the Ap p ly button.
6. Edit the InfiniBand-specific settings by referring to the Configuring the InfiniBand Tab description
below .

14 1

Deployment G uide

Fig u re 9 .15. Ed it in g t h e n ewly creat ed In f in iB an d co n n ect io n 1
Pro ced u re 9 .14 . Ed it in g an Exist in g In f in iB an d C o n n ect io n
Follow these steps to edit an existing InfiniBand connection.
1. Right-click on the N et wo rkMan ag er applet icon in the Notification Area and click Ed it
C o n n ect io n s. The N et wo rk C o n n ect io n s window appears.
2. Select the connection you wish to edit and click the Ed it button.
3. Select the In f in iB an d tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Three settings in the Ed it in g dialog are common to all connection types:
C o n n ect io n n ame — Enter a descriptive name for your network connection. This name
will be used to list this connection in the In f in iB an d section of the N et wo rk
C o n n ect io n s window.
C o n n ect au t o mat ically — Check this box if you want N et wo rkMan ag er to autoconnect to this connection when it is available. See Section 9.2.3, “ Connecting to a
Network Automatically” for more information.
Availab le t o all u sers — Check this box to create a connection available to all users on
the system. Changing this setting may require ro o t privileges. See Section 9.2.4, “ User
and System Connections” for details.
5. Edit the InfiniBand-specific settings by referring to the Configuring the InfiniBand Tab description

14 2

⁠Chapt er 9 . Net workManager

below .

Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your InfiniBand connection, click the Ap p ly button and
N et wo rkMan ag er will immediately save your customized configuration. Given a correct
configuration, you can connect to your new or customized connection by selecting it from the
N et wo rkMan ag er Notification Area applet. See Section 9.2.1, “ Connecting to a Network” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the N et wo rk C o n n ect io n s
window and clicking Ed it to return to the Ed it in g dialog.
Then, to configure:
IPv4 settings for the connection, click the IPv4 Set t in g s tab and proceed to Section 9.3.9.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IPv6 Set t in g s tab and proceed to Section 9.3.9.5,
“ Configuring IPv6 Settings” .

Co nfiguring t he InfiniBand T ab
If you have already added a new InfiniBand connection (see Procedure 9.13, “ Adding a New
InfiniBand Connection” for instructions), you can edit the In f in iB an d tab to set the parent interface
and the InfiniBand ID .
T ran sp o rt mo d e
D atagram or Connected mode can be selected from the drop-down list. Select the same
mode the rest of your IPoIB network is using.
D evice MAC ad d ress
The MAC address of the InfiniBand capable device to be used for the InfiniBand network
traffic.This hardware address field will be pre-filled if you have InfiniBand hardware
installed.
MT U
Optionally sets a Maximum Transmission Unit (MTU) size to be used for packets to be sent
over the InfiniBand connection.

9.3.9. Configuring Connect ion Set t ings
9 .3.9 .1 . Co nfiguring 8 0 2 .1 X Se curit y
802.1X security is the name of the IEEE standard for port-based Network Access Control (PNAC).
Simply put, 802.1X security is a way of defining a logical network out of a physical one. All clients who
want to join the logical network must authenticate with the server (a router, for example) using the
correct 802.1X authentication method.
802.1X security is most often associated with securing wireless networks (WLANs), but can also be
used to prevent intruders with physical access to the network (LAN) from gaining entry. In the past,
D HCP servers were configured not to lease IP addresses to unauthorized users, but for various
reasons this practice is both impractical and insecure, and thus is no longer recommended. Instead,
802.1X security is used to ensure a logically-secure network through port-based authentication.

14 3

Deployment G uide

802.1X provides a framework for WLAN and LAN access control and serves as an envelope for
carrying one of the Extensible Authentication Protocol (EAP) types. An EAP type is a protocol that
defines how WLAN security is achieved on the network.
You can configure 802.1X security for a wired or wireless connection type by opening the N et wo rk
C o n n ect io n s window (see Section 9.2.2, “ Configuring New and Editing Existing Connections” ) and
following the applicable procedure:
Pro ced u re 9 .15. Fo r a wired co n n ect io n ...
1. Either click Ad d , select a new network connection for which you want to configure 802.1X
security and then click C reat e, or select an existing connection and click Ed it .
2. Then select the 802.1X Secu rit y tab and check the U se 802.1X secu rit y f o r t h is
co n n ect io n checkbox to enable settings configuration.
3. Proceed to Section 9.3.9.1.1, “ Configuring TLS (Transport Layer Security) Settings”
Pro ced u re 9 .16 . Fo r a wireless co n n ect io n ...
1. Either click on Ad d , select a new network connection for which you want to configure 802.1X
security and then click C reat e, or select an existing connection and click Ed it .
2. Select the Wireless Secu rit y tab.
3. Then click the Secu rit y dropdown and choose one of the following security methods: LEAP,
D yn amic WEP ( 802.1X) , or WPA & WPA2 En t erp rise.
4. See Section 9.3.9.1.1, “ Configuring TLS (Transport Layer Security) Settings” for descriptions
of which EAP types correspond to your selection in the Secu rit y dropdown.
9 .3.9 .1.1. C o n f ig u rin g T LS ( T ran sp o rt Layer Secu rit y) Set t in g s
With Transport Layer Security, the client and server mutually authenticate using the TLS protocol.
The server demonstrates that it holds a digital certificate, the client proves its own identity using its
client-side certificate, and key information is exchanged. Once authentication is complete, the TLS
tunnel is no longer used. Instead, the client and server use the exchanged keys to encrypt data using
AES, TKIP or WEP.
The fact that certificates must be distributed to all clients who want to authenticate means that the
EAP-TLS authentication method is very strong, but also more complicated to set up. Using TLS
security requires the overhead of a public key infrastructure (PKI) to manage certificates. The benefit
of using TLS security is that a compromised password does not allow access to the (W)LAN: an
intruder must also have access to the authenticating client's private key.
N et wo rkMan ag er does not determine the version of TLS supported. N et wo rkMan ag er gathers the
parameters entered by the user and passes them to the daemon, wp a_su p p lican t , that handles the
procedure. It in turn uses OpenSSL to establish the TLS tunnel. OpenSSL itself negotiates the
SSL/TLS protocol version. It uses the highest version both ends support.
Id en t it y
Identity string for EAP authentication methods, such as a user name or login name.
U ser cert if icat e
Click to browse for, and select, a user's certificate.
C A cert if icat e

14 4

⁠Chapt er 9 . Net workManager

Click to browse for, and select, a Certificate Authority's certificate.
Privat e key
Click to browse for, and select, a user's private key file. Note that the key must be password
protected.
Privat e key p asswo rd
Enter the user password corresponding to the user's private key.
9 .3.9 .1.2. C o n f ig u rin g T u n n eled T LS Set t in g s
An o n ymo u s id en t it y
This value is used as the unencrypted identity.
C A cert if icat e
Click to browse for, and select, a Certificate Authority's certificate.
In n er au t h en t icat io n
PAP — Password Authentication Protocol.
MSC H AP — Challenge Handshake Authentication Protocol.
MSC H APv2 — Microsoft Challenge Handshake Authentication Protocol version 2.
C H AP — Challenge Handshake Authentication Protocol.
U sern ame
Enter the user name to be used in the authentication process.
Passwo rd
Enter the password to be used in the authentication process.
9 .3.9 .1.3. C o n f ig u rin g Pro t ect ed EAP ( PEAP) Set t in g s
An o n ymo u s Id en t it y
This value is used as the unencrypted identity.
C A cert if icat e
Click to browse for, and select, a Certificate Authority's certificate.
PEAP versio n
The version of Protected EAP to use. Automatic, 0 or 1.
In n er au t h en t icat io n
MSC H APv2 — Microsoft Challenge Handshake Authentication Protocol version 2.
MD 5 — Message D igest 5, a cryptographic hash function.
G T C — Generic Token Card.
U sern ame

14 5

Deployment G uide

Enter the user name to be used in the authentication process.
Passwo rd
Enter the password to be used in the authentication process.

9 .3.9 .2 . Co nfiguring Wire le ss Se curit y
Secu rit y
N o n e — D o not encrypt the Wi-Fi connection.
WEP 4 0/128- b it K ey — Wired Equivalent Privacy (WEP), from the IEEE 802.11 standard.
Uses a single pre-shared key (PSK).
WEP 128- b it Passp h rase — An MD 5 hash of the passphrase will be used to derive a
WEP key.
LEAP — Lightweight Extensible Authentication Protocol, from Cisco Systems.
D yn amic WEP ( 802.1X) — WEP keys are changed dynamically.
WPA & WPA2 Perso n al — Wi-Fi Protected Access (WPA), from the draft IEEE 802.11i
standard. A replacement for WEP. Wi-Fi Protected Access II (WPA2), from the 802.11i-2004
standard. Personal mode uses a pre-shared key (WPA-PSK).
WPA & WPA2 En t erp rise — WPA for use with a RAD IUS authentication server to provide
IEEE 802.1X network access control.
Passwo rd
Enter the password to be used in the authentication process.

Prevent Roaming On T he Same Access Point
In the case of WPA and WPA2 (Personal and Enterprise), an option to select between Auto,
WPA and WPA2 has been added. This option is intended for use with an access point that is
offering both WPA and WPA2. Select one of the protocols if you would like to prevent roaming
between the two protocols. Roaming between WPA and WPA2 on the same access point can
cause loss of service.

14 6

⁠Chapt er 9 . Net workManager

Fig u re 9 .16 . Ed it in g t h e Wireless Secu rit y t ab an d select in g t h e WPA p ro t o co l

9 .3.9 .3. Co nfiguring PPP (Po int -t o -Po int ) Se t t ings
C o n f ig u re Met h o d s
U se p o in t - t o - p o in t en cryp t io n ( MPPE)
Microsoft Point-To-Point Encryption protocol (RFC 3078).
Allo w B SD d at a co mp ressio n
PPP BSD Compression Protocol (RFC 1977).
Allo w D ef lat e d at a co mp ressio n
PPP D eflate Protocol (RFC 1979).
U se T C P h ead er co mp ressio n
Compressing TCP/IP Headers for Low-Speed Serial Links (RFC 1144).
Sen d PPP ech o p acket s
LCP Echo-Request and Echo-Reply Codes for loopback tests (RFC 1661).

9 .3.9 .4 . Co nfiguring IPv4 Se t t ings

14 7

Deployment G uide

Fig u re 9 .17. Ed it in g t h e IPv4 Set t in g s T ab
The IPv4 Set t in g s tab allows you to configure the method by which you connect to the Internet and
enter IP address, route, and D NS information as required. The IPv4 Set t in g s tab is available when
you create and modify one of the following connection types: wired, wireless, mobile broadband, VPN
or D SL.
If you are using D HCP to obtain a dynamic IP address from a D HCP server, you can set Met h o d to
Au t o mat ic ( D H C P) .
Set t in g t h e Met h o d

Availab le IPv4 Met h o d s b y C o n n ect io n T yp e
When you click the Met h o d dropdown menu, depending on the type of connection you are
configuring, you are able to select one of the following IPv4 connection methods. All of the methods
are listed here according to which connection type or types they are associated with.
Met h o d
Au t o mat ic ( D H C P) — Choose this option if the network you are connecting to uses a
D HCP server to assign IP addresses. You do not need to fill in the D H C P clien t ID field.

14 8

⁠Chapt er 9 . Net workManager

Au t o mat ic ( D H C P) ad d resses o n ly — Choose this option if the network you are
connecting to uses a D HCP server to assign IP addresses but you want to assign D NS
servers manually.
Lin k- Lo cal O n ly — Choose this option if the network you are connecting to does not have
a D HCP server and you do not want to assign IP addresses manually. Random addresses
will be selected as per RFC 3927.
Sh ared t o o t h er co mp u t ers — Choose this option if the interface you are configuring is
for sharing an Internet or WAN connection.
Wired , Wireless an d D SL C o n n ect io n Met h o d s
Man u al — Choose this option if the network you are connecting to does not have a D HCP
server and you want to assign IP addresses manually.
Mo b ile B ro ad b an d C o n n ect io n Met h o d s
Au t o mat ic ( PPP) — Choose this option if the network you are connecting to uses a D HCP
server to assign IP addresses.
Au t o mat ic ( PPP) ad d resses o n ly — Choose this option if the network you are
connecting to uses a D HCP server to assign IP addresses but you want to assign D NS
servers manually.
VPN C o n n ect io n Met h o d s
Au t o mat ic ( VPN ) — Choose this option if the network you are connecting to uses a D HCP
server to assign IP addresses.
Au t o mat ic ( VPN ) ad d resses o n ly — Choose this option if the network you are
connecting to uses a D HCP server to assign IP addresses but you want to assign D NS
servers manually.
D SL C o n n ect io n Met h o d s
Au t o mat ic ( PPPo E) — Choose this option if the network you are connecting to uses a
D HCP server to assign IP addresses.
Au t o mat ic ( PPPo E) ad d resses o n ly — Choose this option if the network you are
connecting to uses a D HCP server to assign IP addresses but you want to assign D NS
servers manually.
PPPo E Sp ecif ic C o n f ig u rat io n St ep s
If more than one NIC is installed, and PPPoE will only be run over one NIC but not the other, then for
correct PPPoE operation it is also necessary to lock the connection to the specific Ethernet device
PPPoE is supposed to be run over. To lock the connection to one specific NIC, do o n e of the
following:
Enter the MAC address in n m- co n n ect io n - ed it o r for that connection. Optionally select
C o n n ect au t o mat ically and Availab le t o all u sers to make the connection come up without
requiring user login after system start.
Set the hardware-address in the [802-3-ethernet] section in the appropriate file for that connection
in /et c/N et wo rkMan ag er/syst em- co n n ect io n s/ as follows:
[802-3-ethernet]
mac-address=00:11:22:33:44:55

14 9

Deployment G uide

Mere presence of the file in /et c/N et wo rkMan ag er/syst em- co n n ect io n s/ means that it is
“ available to all users” . Ensure that au t o co n n ect = t ru e appears in the [connection] section for
the connection to be brought up without requiring user login after system start.
For information on configuring static routes for the network connection, go to Section 9.3.9.6,
“ Configuring Routes” .

9 .3.9 .5 . Co nfiguring IPv6 Se t t ings
Met h o d
Ig n o re — Choose this option if you want to disable IPv6 settings.
Au t o mat ic — Choose this option if the network you are connecting to uses a D HCP server
to assign IP addresses.
Au t o mat ic, ad d resses o n ly — Choose this option if the network you are connecting to
uses a D HCP server to assign IP addresses but you want to assign D NS servers manually.
Man u al — Choose this option if the network you are connecting to does not have a D HCP
server and you want to assign IP addresses manually.
Lin k- Lo cal O n ly — Choose this option if the network you are connecting to does not have
a D HCP server and you do not want to assign IP addresses manually. Random addresses
will be selected as per RFC 4862.
Sh ared t o o t h er co mp u t ers — Choose this option if the interface you are configuring is
for sharing an Internet or WAN connection.
Ad d resses
D N S servers — Enter a comma separated list of D NS servers.
Search d o main s — Enter a comma separated list of domain controllers.
For information on configuring static routes for the network connection, go to Section 9.3.9.6,
“ Configuring Routes” .

9 .3.9 .6 . Co nfiguring Ro ut e s
A host's routing table will be automatically populated with routes to directly connected networks. The
routes are learned by observing the network interfaces when they are “ up” . This section is for
entering static routes to networks or hosts which can be reached by traversing an intermediate
network or connection, such as a VPN or leased line.

150

⁠Chapt er 9 . Net workManager

Fig u re 9 .18. C o n f ig u rin g st at ic n et wo rk ro u t es
Ad d resses
Ad d ress — The IP address of a network, sub-net or host.
N et mask — The netmask or prefix length of the IP address just entered.
G at eway — The IP address of the gateway leading to the network, sub-net or host.
Met ric — A network cost, that is to say a preference value to give to this route. Lower
values will be preferred over higher values.
Ig n o re au t o mat ically o b t ain ed ro u t es
Select this checkbox to only use manually entered routes for this connection.
U se t h is co n n ect io n o n ly f o r reso u rces o n it s n et wo rk
Select this checkbox to prevent the connection from becoming the default route. Typical
examples are where a connection is a VPN or a leased line to a head office and you do not
want any Internet bound traffic to pass over the connection. Selecting this option means
that only traffic specifically destined for routes learned automatically over the connection or
entered here manually will be routed over the connection.

151

Deployment G uide

Chapter 10. Network Interfaces
Under Red Hat Enterprise Linux, all network communications occur between configured software
interfaces and physical networking devices connected to the system.
The configuration files for network interfaces are located in the /et c/sysco n f ig /n et wo rk- scrip t s/
directory. The scripts used to activate and deactivate these network interfaces are also located here.
Although the number and type of interface files can differ from system to system, there are three
categories of files that exist in this directory:
1. Interface configuration files
2. Interface control scripts
3. Network function files
The files in each of these categories work together to enable various network devices.
This chapter explores the relationship between these files and how they are used.

10.1. Net work Configurat ion Files
Before delving into the interface configuration files, let us first itemize the primary configuration files
used in network configuration. Understanding the role these files play in setting up the network stack
can be helpful when customizing a Red Hat Enterprise Linux system.
The primary network configuration files are as follows:
/et c/h o st s
The main purpose of this file is to resolve host names that cannot be resolved any other
way. It can also be used to resolve host names on small networks with no D N S server.
Regardless of the type of network the computer is on, this file should contain a line
specifying the IP address of the loopback device (127.0.0.1) as lo calh o st .lo cald o main .
For more information, see the h o st s( 5) manual page.
/et c/reso lv.co n f
This file specifies the IP addresses of D N S servers and the search domain. Unless
configured to do otherwise, the network initialization scripts populate this file. For more
information about this file, see the reso lv.co n f ( 5) manual page.
/et c/sysco n f ig /n et wo rk
This file specifies routing and host information for all network interfaces. It is used to
contain directives which are to have global effect and not to be interface specific. For more
information about this file and the directives it accepts, see Section D .1.13,
“ /etc/sysconfig/network” .
/et c/sysco n f ig /n et wo rk- scrip t s/if cf g - interface-name
For each network interface, there is a corresponding interface configuration script. Each of
these files provide information specific to a particular network interface. See Section 10.2,
“ Interface Configuration Files” for more information on this type of file and the directives it
accepts.

152

⁠Chapt er 1 0 . Net work Int erfaces

Network interface names
Network interface names may be different on different hardware types. See Appendix A,
Consistent Network Device Naming for more information.

T he /etc/sysconfig/networking/ directory
The /et c/sysco n f ig /n et wo rkin g / directory is used by the now deprecated N et wo rk
Ad min ist rat io n T o o l (syst em- co n f ig - n et wo rk). Its contents should n o t be edited
manually. Using only one method for network configuration is strongly encouraged, due to the
risk of configuration deletion. For more information about configuring network interfaces using
graphical configuration tools, see Chapter 9, NetworkManager.

10.2. Int erface Configurat ion Files
Interface configuration files control the software interfaces for individual network devices. As the
system boots, it uses these files to determine what interfaces to bring up and how to configure them.
These files are usually named if cf g - name, where name refers to the name of the device that the
configuration file controls.

10.2.1. Et hernet Int erfaces
One of the most common interface files is /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h 0, which
controls the first Ethernet network interface card or NIC in the system. In a system with multiple NICs,
there are multiple if cf g - et h X files (where X is a unique number corresponding to a specific
interface). Because each device has its own configuration file, an administrator can control how
each interface functions individually.
The following is a sample if cf g - et h 0 file for a system using a fixed IP address:
D EVICE=eth0
BOOTPROTO=none
ONBOOT=yes
NETMASK=255.255.255.0
IPAD D R=10.0.1.27
USERCTL=no
The values required in an interface configuration file can change based on other values. For
example, the if cf g - et h 0 file for an interface using D H C P looks different because IP information is
provided by the D H C P server:
D EVICE=eth0
BOOTPROTO=dhcp
ONBOOT=yes
N et wo rkMan ag er is graphical configuration tool which provides an easy way to make changes to
the various network interface configuration files (see Chapter 9, NetworkManager for detailed
instructions on using this tool).
However, it is also possible to manually edit the configuration files for a given network interface.

153

Deployment G uide

Below is a listing of the configurable parameters in an Ethernet interface configuration file:
B O N D IN G _O PT S= parameters
sets the configuration parameters for the bonding device, and is used in
/et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d N (see Section 10.2.4, “ Channel Bonding
Interfaces” ). These parameters are identical to those used for bonding devices in
/sys/class/n et /bonding_device/b o n d in g , and the module parameters for the bonding
driver as described in bonding Module Directives.
This configuration method is used so that multiple bonding devices can have different
configurations. In Red Hat Enterprise Linux 6, place all interface-specific bonding options
after the B O N D IN G _O PT S directive in if cf g - name files. See Where to specify bonding
module parameters for more information.
B O O T PR O T O = protocol
where protocol is one of the following:
n o n e — No boot-time protocol should be used.
b o o t p — The B O O T P protocol should be used.
d h cp — The D H C P protocol should be used.
B R O AD C AST = address
where address is the broadcast address. This directive is deprecated, as the value is
calculated automatically with ip calc.
D EVIC E= name
where name is the name of the physical device (except for dynamically-allocated PPP
devices where it is the logical name).
D H C P_H O ST N AME= name
where name is a short host name to be sent to the D H C P server. Use this option only if the
D H C P server requires the client to specify a host name before receiving an IP address.
D H C PV6 C = answer
where answer is one of the following:
yes — Use D H C P to obtain an IPv6 address for this interface.
n o — D o not use D H C P to obtain an IPv6 address for this interface. This is the default
value.
An IPv6 link-local address will still be assigned by default. The link-local address is based
on the MAC address of the interface as per RFC 4862.
D H C PV6 C _O PT IO N S= answer
where answer is one of the following:
- P — Enable IPv6 prefix delegation.
- S — Use D H C P to obtain stateless configuration only, not addresses, for this interface.
- N — Restore normal operation after using the - T or - P options.

154

⁠Chapt er 1 0 . Net work Int erfaces

- T — Use D H C P to obtain a temporary IPv6 address for this interface.
- D — Override the default when selecting the type of DHCP Unique Identifier (D UID ) to
use.
By default, the D H C P client (dhclient) creates a DHCP Unique Identifier (D UID ) based on
the link-layer address (D UID -LL) if it is running in stateless mode (with the - S option, to
not request an address), or it creates an identifier based on the link-layer address plus a
timestamp (D UID -LLT) if it is running in stateful mode (without - S, requesting an
address). The - D option overrides this default, with a value of either LL or LLT .
D N S{1,2}= address
where address is a name server address to be placed in /et c/reso lv.co n f provided that the
PEER D N S directive is not set to n o .
ET H T O O L_O PT S= options
where options are any device-specific options supported by et h t o o l. For example, if you
wanted to force 100Mb, full duplex:
ETHTOOL_OPTS=" autoneg off speed 100 duplex full"
Instead of a custom initscript, use ET H T O O L_O PT S to set the interface speed and duplex
settings. Custom initscripts run outside of the network init script lead to unpredictable
results during a post-boot network service restart.

Set “autoneg off” before changing speed or duplex
settings
Changing speed or duplex settings almost always requires disabling autonegotiation with the au t o n eg o f f option. This option needs to be stated first, as the
option entries are order-dependent.
See Section 10.8, “ Ethtool” for more et h t o o l options.
H O T PLU G = answer
where answer is one of the following:
yes — This device should be activated when it is hot-plugged (this is the default option).
n o — This device should not be activated when it is hot-plugged.
The H O T PLU G = n o option can be used to prevent a channel bonding interface from being
activated when a bonding kernel module is loaded.
See Section 10.2.4, “ Channel Bonding Interfaces” for more information about channel
bonding interfaces.
H WAD D R = MAC-address
where MAC-address is the hardware address of the Ethernet device in the form
AA:BB:CC:DD:EE:FF. This directive must be used in machines containing more than one NIC
to ensure that the interfaces are assigned the correct device names regardless of the
configured load order for each NIC's module. This directive should n o t be used in

155

Deployment G uide

conjunction with MAC AD D R .

Note
Persistent device names are now handled by /et c/u d ev/ru les.d /70- p ersist en t n et .ru les.
H WAD D R must not be used with System z network devices.
See Section 25.3.3, " Mapping subchannels and network device names" , in the
Red Hat Enterprise Linux 6 Installation Guide.

IPAD D R = address
where address is the IPv4 address.
IPV6 AD D R = address
where address is the first static, or primary, IPv6 address on an interface.
The format is Address/Prefix-length. If no prefix length is specified, /6 4 is assumed. Note
that this setting depends on IPV6 IN IT being enabled.
IPV6 AD D R _SEC O N D AR IES= address
where address is one or more, space separated, additional IPv6 addresses.
The format is Address/Prefix-length. If no prefix length is specified, /6 4 is assumed. Note
that this setting depends on IPV6 IN IT being enabled.
IPV6 IN IT = answer
where answer is one of the following:
yes — Initialize this interface for IPv6 addressing.
n o — D o not initialize this interface for IPv6 addressing. This is the default value.
This setting is required for IPv6 static and D H C P assignment of IPv6 addresses. It does
not affect IPv6 Stateless Address Autoconfiguration (SLAAC) as per RFC 4862.
See Section D .1.13, “ /etc/sysconfig/network” for information on disabling IPv6 .
IPV6 _AU T O C O N F= answer
where answer is one of the following:
yes — Enable IPv6 autoconf configuration for this interface.
n o — D isable IPv6 autoconf configuration for this interface.
If enabled, an IPv6 address will be requested using Neighbor Discovery (ND ) from a router
running the rad vd daemon.
Note that the default value of IPV6 _AU T O C O N F depends on IPV6 FO R WAR D IN G as
follows:
If IPV6 FO R WAR D IN G =yes, then IPV6 _AU T O C O N F will default to n o .

156

⁠Chapt er 1 0 . Net work Int erfaces

If IPV6 FO R WAR D IN G =n o , then IPV6 _AU T O C O N F will default to yes and
IPV6 _R O U T ER has no effect.
IPV6 _MT U = value
where value is an optional dedicated MTU for this interface.
IPV6 _PR IVAC Y= rfc3041
where rfc3041 optionally sets this interface to support RFC 3041 Privacy Extensions for
Stateless Address Autoconfiguration in IPv6. Note that this setting depends on IPV6 IN IT
option being enabled.
The default is for RFC 3041 support to be disabled. Stateless Autoconfiguration will derive
addresses based on the MAC address, when available, using the modified EU I- 6 4 method.
The address is appended to a prefix but as the address is normally derived from the MAC
address it is globally unique even when the prefix changes. In the case of a link-local
address the prefix is f e80::/6 4 as per RFC 2462 IPv6 Stateless Address Autoconfiguration.
LIN K D ELAY= time
where time is the number of seconds to wait for link negotiation before configuring the
device. The default is 5 secs. D elays in link negotiation, caused by ST P for example, can
be overcome by increasing this value.
MAC AD D R = MAC-address
where MAC-address is the hardware address of the Ethernet device in the form
AA:BB:CC:DD:EE:FF.
This directive is used to assign a MAC address to an interface, overriding the one assigned
to the physical NIC. This directive should n o t be used in conjunction with the H WAD D R
directive.
MAST ER = bond-interface
where bond-interface is the channel bonding interface to which the Ethernet interface is
linked.
This directive is used in conjunction with the SLAVE directive.
See Section 10.2.4, “ Channel Bonding Interfaces” for more information about channel
bonding interfaces.
N ET MASK = mask
where mask is the netmask value.
N ET WO R K = address
where address is the network address. This directive is deprecated, as the value is
calculated automatically with ip calc.
N M_C O N T R O LLED = answer
where answer is one of the following:
yes — N et wo rkMan ag er is permitted to configure this device. This is the default
behavior and can be omitted.
n o — N et wo rkMan ag er is not permitted to configure this device.

157

Deployment G uide

Note
The N M_C O N T R O LLED directive is now, as of Red Hat Enterprise Linux 6.3,
dependent on the N M_B O N D _VLAN _EN AB LED directive in
/et c/sysco n f ig /n et wo rk. If and only if that directive is present and is one of yes, y,
or t ru e, will N et wo rkMan ag er detect and manage bonding and VLAN interfaces.
O N B O O T = answer
where answer is one of the following:
yes — This device should be activated at boot-time.
n o — This device should not be activated at boot-time.
PEER D N S= answer
where answer is one of the following:
yes — Modify /et c/reso lv.co n f if the D N S directive is set, if using D H C P, or if using
Microsoft's RFC 1877 IPC P extensions with PPP. In all cases yes is the default.
n o — D o not modify /et c/reso lv.co n f .
SLAVE= answer
where answer is one of the following:
yes — This device is controlled by the channel bonding interface specified in the
MAST ER directive.
n o — This device is not controlled by the channel bonding interface specified in the
MAST ER directive.
This directive is used in conjunction with the MAST ER directive.
See Section 10.2.4, “ Channel Bonding Interfaces” for more about channel bonding
interfaces.
SR C AD D R = address
where address is the specified source IP address for outgoing packets.
U SER C T L= answer
where answer is one of the following:
yes — Non-ro o t users are allowed to control this device.
n o — Non-ro o t users are not allowed to control this device.

10.2.2. Specific ifcfg Opt ions for Linux on Syst em z
SU B C H AN N ELS= , , 
where , , and  are the three
device bus ID s representing a network device.

158

⁠Chapt er 1 0 . Net work Int erfaces

PO R T N AME= myname;
where myname is the Open Systems Adapter (OSA) portname or LAN Channel Station (LCS)
portnumber.
C T C PR O T = answer
where answer is one of the following:
0 — Compatibility mode, TCP/IP for Virtual Machines (used with non-Linux peers other
than IBM S/390 and IBM System z operating systems). This is the default mode.
1 — Extended mode, used for Linux-to-Linux Peers.
3 — Compatibility mode for S/390 and IBM System z operating systems.
This directive is used in conjunction with the NETTYPE directive. It specifies the CTC
protocol for NETTYPE='ctc'. The default is 0.
O PT IO N = 'answer'
where 'answer' is a quoted string of any valid sysfs attributes and their value. The Red Hat
Enterprise Linux installer currently uses this to configure the layer mode, (layer2), and the
relative port number, (portno), of QETH devices. For example:
OPTIONS='layer2=1 portno=0'

10.2.3. Required ifcfg Opt ions for Linux on Syst em z
N ET T YPE= answer
where answer is one of the following:
ct c — Channel-to-Channel communication. For point-to-point TCP/IP or TTY.
lcs — LAN Channel Station (LCS).
q et h — QETH (QD IO Ethernet). This is the default network interface. It is the preferred
installation method for supporting real or virtual OSA cards and HiperSockets devices.

10.2.4 . Channel Bonding Int erfaces
Red Hat Enterprise Linux allows administrators to bind multiple network interfaces together into a
single channel using the b o n d in g kernel module and a special network interface called a channel
bonding interface. Channel bonding enables two or more network interfaces to act as one,
simultaneously increasing the bandwidth and providing redundancy.

Warning
The use of direct cable connections without network switches is not supported for bonding.
The failover mechanisms described here will not work as expected without the presence of
network switches. See the Red Hat Knowledgebase article Why is bonding in not supported with
direct connection using crossover cables? for more information.

159

Deployment G uide

Note
The active-backup, balance-tlb and balance-alb modes do not require any specific
configuration of the switch. Other bonding modes require configuring the switch to aggregate
the links. For example, a Cisco switch requires EtherChannel for Modes 0, 2, and 3, but for
Mode 4 LACP and EtherChannel are required. See the documentation supplied with your
switch and the b o n d in g .t xt file in the kernel-doc package (see Section 29.9, “ Additional
Resources” ).

1 0 .2 .4 .1 . Che ck if Bo nding Ke rne l Mo dule is Inst alle d
In Red Hat Enterprise Linux 6, the bonding module is not loaded by default. You can load the module
by issuing the following command as ro o t :
~]# mo d p ro b e - - f irst - t ime b o n d in g
No visual output indicates the module was not running and has now been loaded. This activation
will not persist across system restarts. See Section 29.7, “ Persistent Module Loading” for an
explanation of persistent module loading. Note that given a correct configuration file using the
B O N D IN G _O PT S directive, the bonding module will be loaded as required and therefore does not
need to be loaded separately.
To display information about the module, issue the following command:
~]$ mo d in f o b o n d in g
See the mo d p ro b e( 8) man page for more command options and see Chapter 29, Working with Kernel
Modules for information on loading and unloading modules.

1 0 .2 .4 .2 . Cre at e a Channe l Bo nding Int e rface
To create a channel bonding interface, create a file in the /et c/sysco n f ig /n et wo rk- scrip t s/
directory called if cf g - b o n d N, replacing N with the number for the interface, such as 0.
The contents of the file can be identical to whatever type of interface is getting bonded, such as an
Ethernet interface. The only difference is that the D EVIC E directive is b o n d N, replacing N with the
number for the interface. The N M_C O N T R O LLED directive can be added to prevent
N et wo rkMan ag er from configuring this device.

Examp le 10.1. Examp le if cf g - b o n d 0 in t erf ace co n f ig u rat io n f ile
The following is an example of a channel bonding interface configuration file:
D EVICE=bond0
IPAD D R=192.168.1.1
NETMASK=255.255.255.0
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
NM_CONTROLLED =no
BOND ING_OPTS=" bonding parameters separated by spaces"

160

⁠Chapt er 1 0 . Net work Int erfaces

The MAC address of the bond will be taken from the first interface to be enslaved. It can also be
specified using the HWAD D R directive if required. If you want N et wo rkMan ag er to control this
interface, remove the N M_C O N T R O LLED = n o directive, or set it to yes, and add T YPE= B o n d
and B O N D IN G _MAST ER = yes.
After the channel bonding interface is created, the network interfaces to be bound together must be
configured by adding the MAST ER and SLAVE directives to their configuration files. The
configuration files for each of the channel-bonded interfaces can be nearly identical.

Examp le 10.2. Examp le if cf g - et h X b o n d ed in t erf ace co n f ig u rat io n f ile
If two Ethernet interfaces are being channel bonded, both et h 0 and et h 1 can be as follows:
D EVICE=ethX
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no
NM_CONTROLLED =no
In this example, replace X with the numerical value for the interface.
Once the interfaces have been configured, restart the network service to bring the bond up. As ro o t ,
issue the following command:
~]# service n et wo rk rest art
To view the status of a bond, view the /p ro c/ file by issuing a command in the following format:
cat /proc/net/bonding/bondN
For example:
~]$ cat /p ro c/n et /b o n d in g /b o n d 0
Ethernet Channel Bonding D river: v3.6.0 (September 26, 2009)
Bonding Mode: load balancing (round-robin)
MII Status: down
MII Polling Interval (ms): 0
Up D elay (ms): 0
D own D elay (ms): 0
For further instructions and advice on configuring the bonding module and to view the list of
bonding parameters, see Section 29.8.1, “ Using Channel Bonding” .
Support for bonding was added to N et wo rkMan ag er in Red Hat Enterprise Linux 6.3. See
Section 10.2.1, “ Ethernet Interfaces” for an explanation of N M_C O N T R O LLED and the
N M_B O N D _VLAN _EN AB LED directive.

161

Deployment G uide

Where to specify bonding module parameters
In Red Hat Enterprise Linux 6, interface-specific parameters for the bonding kernel module
must be specified as a space-separated list in the B O N D IN G _O PT S= "bonding parameters"
directive in the if cf g - b o n d N interface file. D o n o t specify options specific to a bond in
/et c/mo d p ro b e.d /bonding.co n f , or in the deprecated /et c/mo d p ro b e.co n f file.
The max_b o n d s parameter is not interface specific and therefore, if required, should be
specified in /et c/mo d p ro b e.d /b o n d in g .co n f as follows:
options bonding max_bonds=1
However, the max_b o n d s parameter should n o t be set when using if cf g - b o n d N files with
the B O N D IN G _O PT S directive as this directive will cause the network scripts to create the
bond interfaces as required.
Note that any changes to /et c/mo d p ro b e.d /b o n d in g .co n f will not take effect until the
module is next loaded. A running module must first be unloaded. See Chapter 29, Working with
Kernel Modules for more information on loading and unloading modules.

10.2.4 .2.1. C reat in g Mu lt ip le B o n d s
In Red Hat Enterprise Linux 6, for each bond a channel bonding interface is created including the
B O N D IN G _O PT S directive. This configuration method is used so that multiple bonding devices can
have different configurations. To create multiple channel bonding interfaces, proceed as follows:
Create multiple if cf g - b o n d N files with the B O N D IN G _O PT S directive; this directive will cause
the network scripts to create the bond interfaces as required.
Create, or edit existing, interface configuration files to be bonded and include the SLAVE
directive.
Assign the interfaces to be bonded, the slave interfaces, to the channel bonding interfaces by
means of the MAST ER directive.

Examp le 10.3. Examp le mu lt ip le if cf g - b o n d N in t erf ace co n f ig u rat io n f iles
The following is an example of a channel bonding interface configuration file:
D EVICE=bondN
IPAD D R=192.168.1.1
NETMASK=255.255.255.0
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
NM_CONTROLLED =no
BOND ING_OPTS=" bonding parameters separated by spaces"
In this example, replace N with the number for the bond interface. For example, to create two bonds
create two configuration files, if cf g - b o n d 0 and if cf g - b o n d 1.
Create the interfaces to be bonded as per Example 10.2, “ Example ifcfg-ethX bonded interface

162

⁠Chapt er 1 0 . Net work Int erfaces

configuration file” and assign them to the bond interfaces as required using the MAST ER = b o n d N
directive. For example, continuing on from the example above, if two interfaces per bond are required,
then for two bonds create four interface configuration files and assign the first two using
MAST ER = b o n d 0 and the next two using MAST ER = b o n d 1.

10.2.5. Configuring a VLAN over a Bond
This section will show configuring a VLAN over a bond consisting of two Ethernet links between a
server and an Ethernet switch. The switch has a second bond to another server. Only the
configuration for the first server will be shown as the other is essentially the same apart from the IP
addresses.

Warning
The use of direct cable connections without network switches is not supported for bonding.
The failover mechanisms described here will not work as expected without the presence of
network switches. See the Red Hat Knowledgebase article Why is bonding in not supported with
direct connection using crossover cables? for more information.

Note
The active-backup, balance-tlb and balance-alb modes do not require any specific
configuration of the switch. Other bonding modes require configuring the switch to aggregate
the links. For example, a Cisco switch requires EtherChannel for Modes 0, 2, and 3, but for
Mode 4 LACP and EtherChannel are required. See the documentation supplied with your
switch and the b o n d in g .t xt file in the kernel-doc package (see Section 29.9, “ Additional
Resources” ).
Check the available interfaces on the server:
~]$ ip ad d r
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast state D OWN qlen 1000
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
3: eth1:  mtu 1500 qdisc pfifo_fast state D OWN qlen 1000
link/ether 52:54:00:f6:63:9a brd ff:ff:ff:ff:ff:ff
Pro ced u re 10.1. C o n f ig u rin g t h e In t erf aces o n t h e Server
1. Configure a slave interface using et h 0:
~]# vi /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h 0
NAME=bond0-slave0
D EVICE=eth0
TYPE=Ethernet
BOOTPROTO=none

163

Deployment G uide

ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED =no
The use of the NAME directive is optional. It is for display by a GUI interface, such as n mco n n ect io n - ed it o r and n m- ap p let .
2. Configure a slave interface using et h 1:
~]# vi /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h 1
NAME=bond0-slave1
D EVICE=eth1
TYPE=Ethernet
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
NM_CONTROLLED =no
The use of the NAME directive is optional. It is for display by a GUI interface, such as n mco n n ect io n - ed it o r and n m- ap p let .
3. Configure a channel bonding interface if cf g - b o n d 0:
~]# vi /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0
NAME=bond0
D EVICE=bond0
BOND ING_MASTER=yes
TYPE=Bond
IPAD D R=192.168.100.100
NETMASK=255.255.255.0
ONBOOT=yes
BOOTPROTO=none
BOND ING_OPTS=" mode=active-backup miimon=100"
NM_CONTROLLED =no
The use of the NAME directive is optional. It is for display by a GUI interface, such as n mco n n ect io n - ed it o r and n m- ap p let . In this example MII is used for link monitoring, see the
Section 29.8.1.1, “ Bonding Module D irectives” section for more information on link
monitoring.
4. Check the status of the interfaces on the server:
~]$ ip ad d r
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast state UP
qlen 1000
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
inet6 fe80::5054:ff:fe19:28fe/64 scope link
valid_lft forever preferred_lft forever
3: eth1:  mtu 1500 qdisc pfifo_fast state UP

164

⁠Chapt er 1 0 . Net work Int erfaces

qlen 1000
link/ether 52:54:00:f6:63:9a brd ff:ff:ff:ff:ff:ff
inet6 fe80::5054:ff:fef6:639a/64 scope link
valid_lft forever preferred_lft forever
Pro ced u re 10.2. R eso lvin g C o n f lict s wit h In t erf aces
The interfaces configured as slaves should not have IP addresses assigned to them apart from the
IPv6 link-local addresses (starting f e80). If you have an unexpected IP address, then there may be
another configuration file with ONBOOT set to yes.
1. If this occurs, issue the following command to list all if cf g files that may be causing a
conflict:
~]$ g rep - r "O N B O O T = yes" /et c/sysco n f ig /n et wo rk- scrip t s/ | cu t - f 1 - d ":" |
xarg s g rep - E "IPAD D R |SLAVE"
/etc/sysconfig/network-scripts/ifcfg-lo:IPAD D R=127.0.0.1
The above shows the expected result on a new installation. Any file having both the ONBOOT
directive as well as the IPAD D R or SLAVE directive will be displayed. For example, if the
if cf g - et h 1 file was incorrectly configured, the display might look similar to the following:
~]# g rep - r "O N B O O T = yes" /et c/sysco n f ig /n et wo rk- scrip t s/ | cu t - f 1 - d ":" |
xarg s g rep - E "IPAD D R |SLAVE"
/etc/sysconfig/network-scripts/ifcfg-lo:IPAD D R=127.0.0.1
/etc/sysconfig/network-scripts/ifcfg-eth1:SLAVE=yes
/etc/sysconfig/network-scripts/ifcfg-eth1:IPAD D R=192.168.55.55
2. Any other configuration files found should be moved to a different directory for backup, or
assigned to a different interface by means of the HWAD D R directive. After resolving any
conflict set the interfaces “ down” and “ up” again or restart the network service as ro o t :
~]# service n et wo rk rest art
Shutting down interface bond0:
[ OK ]
Shutting down loopback interface:
[ OK ]
Bringing up loopback interface:
[ OK ]
Bringing up interface bond0: D etermining if ip address 192.168.100.100 is already in
use for device bond0...
[ OK ]
If you are using N et wo rkMan ag er, you might need to restart it at this point to make it forget
the unwanted IP address. As ro o t :
~]# service N et wo rkMan ag er rest art
Pro ced u re 10.3. C h eckin g t h e b o n d o n t h e Server
1. Bring up the bond on the server as ro o t :
~]# if u p /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0
D etermining if ip address 192.168.100.100 is already in use for device bond0...
2. Check the status of the interfaces on the server:

165

Deployment G uide

~]$ ip ad d r
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast
mast er b o n d 0 st at e U P qlen 1000
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
3: eth1:  mtu 1500 qdisc pfifo_fast
mast er b o n d 0 st at e U P qlen 1000
link/ether 52:54:00:f6:63:9a brd ff:ff:ff:ff:ff:ff
4: bond0:  mtu 1500 qdisc
noqueue state UP
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
inet 192.168.100.100/24 brd 192.168.100.255 scope global bond0
inet6 fe80::5054:ff:fe19:28fe/64 scope link
valid_lft forever preferred_lft forever
Notice that et h 0 and et h 1 have mast er b o n d 0 st at e U P and b o n d 0 has status of
MAST ER ,U P.
3. View the bond configuration details:
~]$ cat /p ro c/n et /b o n d in g /b o n d 0
Ethernet Channel Bonding D river: v3.6.0 (September 26, 2009)
Bonding Mode: transmit load balancing
Primary Slave: None
Currently Active Slave: eth0
MII Status: up
MII Polling Interval (ms): 100
Up D elay (ms): 0
D own D elay (ms): 0
Slave Interface: eth0
MII Status: up
Speed: 100 Mbps
D uplex: full
Link Failure Count: 0
Permanent HW addr: 52:54:00:19:28:fe
Slave queue ID : 0
Slave Interface: eth1
MII Status: up
Speed: 100 Mbps
D uplex: full
Link Failure Count: 0
Permanent HW addr: 52:54:00:f6:63:9a
Slave queue ID : 0
4. Check the routes on the server:

166

⁠Chapt er 1 0 . Net work Int erfaces

~]$ ip ro u t e
192.168.100.0/24 dev bond0 proto kernel scope link src 192.168.100.100
169.254.0.0/16 dev bond0 scope link metric 1004
Pro ced u re 10.4 . C o n f ig u rin g t h e VLAN o n t h e Server

Important
At the time of writing, it is important that the bond has slaves and that they are “ up” before
bringing up the VLAN interface. At the time of writing, adding a VLAN interface to a bond
without slaves does not work. In Red Hat Enterprise Linux 6, setting the ONPARENT directive to
yes is important to ensure that the VLAN interface does not attempt to come up before the bond
is up. This is because a VLAN virtual device takes the MAC address of its parent, and when a
NIC is enslaved, the bond changes its MAC address to that NIC's MAC address.

Note
A VLAN slave cannot be configured on a bond with the f ail_o ver_mac= f o llo w option,
because the VLAN virtual device cannot change its MAC address to match the parent's new
MAC address. In such a case, traffic would still be sent with the now incorrect source MAC
address.
Some older network interface cards, loopback interfaces, Wimax cards, and some Infiniband
devices, are said to be VLAN challenged, meaning they cannot support VLANs. This is usually
because the devices cannot cope with VLAN headers and the larger MTU size associated with
VLANs.

1. Create a VLAN interface file b o n d 0.19 2:
~]# vi /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0.19 2
D EVICE=bond0.192
NAME=bond0.192
BOOTPROTO=none
ONPARENT=yes
IPAD D R=192.168.10.1
NETMASK=255.255.255.0
VLAN=yes
NM_CONTROLLED =no
2. Bring up the VLAN interface as ro o t :
~]# if u p /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0.19 2
D etermining if ip address 192.168.10.1 is already in use for device bond0.192...
3. Enabling VLAN tagging on the network switch. Consult the documentation for the switch to
see what configuration is required.
4. Check the status of the interfaces on the server:

167

Deployment G uide

~]# ip ad d r
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0:  mtu 1500 qdisc pfifo_fast
master bond0 state UP qlen 1000
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
3: eth1:  mtu 1500 qdisc pfifo_fast
master bond0 state UP qlen 1000
link/ether 52:54:00:f6:63:9a brd ff:ff:ff:ff:ff:ff
4: bond0:  mtu 1500 qdisc
noqueue state UP
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
inet 192.168.100.100/24 brd 192.168.100.255 scope global bond0
inet6 fe80::5054:ff:fe19:28fe/64 scope link
valid_lft forever preferred_lft forever
5: b o n d 0.19 2@ b o n d 0:  mtu
1500 qdisc noqueue state UP
link/ether 52:54:00:19:28:fe brd ff:ff:ff:ff:ff:ff
inet 192.168.10.1/24 brd 192.168.10.255 scope global bond0.192
inet6 fe80::5054:ff:fe19:28fe/64 scope link
valid_lft forever preferred_lft forever
Notice there is now b o n d 0.19 2@ b o n d 0 in the list of interfaces and the status is
MAST ER ,U P.
5. Check the route on the server:
~]$ ip ro u t e
192.168.100.0/24 dev bond0 proto kernel scope link src 192.168.100.100
192.168.10.0/24 dev b o n d 0.19 2 proto kernel scope link src 192.168.10.1
169.254.0.0/16 dev bond0 scope link metric 1004
169.254.0.0/16 dev bond0.192 scope link metric 1005
Notice there is now a route for the 19 2.16 8.10.0/24 network pointing to the VLAN interface
b o n d 0.19 2.

Co nfiguring t he Se co nd Se rve r
Repeat the configuration steps for the second server, using different IP addresses but from the same
subnets respectively.
Test the bond is up and the network switch is working as expected:
~]$ p in g - c4 19 2.16 8.100.100
PING 192.168.100.100 (192.168.100.100) 56(84) bytes of data.
64 bytes from 192.168.100.100: icmp_seq=1 ttl=64 time=1.35 ms
64 bytes from 192.168.100.100: icmp_seq=2 ttl=64 time=0.214 ms
64 bytes from 192.168.100.100: icmp_seq=3 ttl=64 time=0.383 ms
64 bytes from 192.168.100.100: icmp_seq=4 ttl=64 time=0.396 ms

168

⁠Chapt er 1 0 . Net work Int erfaces

--- 192.168.100.100 ping statistics --4 packets transmitted, 4 received, 0% packet loss, time 3002ms
rtt min/avg/max/mdev = 0.214/0.586/1.353/0.448 ms

T e st ing t he VLAN
To test that the network switch is configured for the VLAN, try to ping the first servers' VLAN interface:
~]# p in g - c2 19 2.16 8.10.1
PING 192.168.10.1 (192.168.10.1) 56(84) bytes of data.
64 bytes from 192.168.10.1: icmp_seq=1 ttl=64 time=0.781 ms
64 bytes from 192.168.10.1: icmp_seq=2 ttl=64 time=0.977 ms
--- 192.168.10.1 ping statistics --2 packets transmitted, 2 received, 0% p acket lo ss, time 1001ms
rtt min/avg/max/mdev = 0.781/0.879/0.977/0.098 ms
No packet loss suggests everything is configured correctly and that the VLAN and underlying
interfaces are “ up” .

Opt io nal St e ps
If required, perform further tests by removing and replacing network cables one at a time to verify
that failover works as expected. Make use of the et h t o o l utility to verify which interface is
connected to which cable. For example:
ethtool - - id en t if y ifname integer
Where integer is the number of times to flash the LED on the network interface.
The bonding module does not support ST P, therefore consider disabling the sending of BPD U
packets from the network switch.
If the system is not linked to the network except over the connection just configured, consider
enabling the switch port to transition directly to sending and receiving. For example on a Cisco
switch, by means of the p o rt f ast command.

10.2.6. Net work Bridge
A network bridge is a Link Layer device which forwards traffic between networks based on MAC
addresses and is therefore also referred to as a Layer 2 device. It makes forwarding decisions based
on tables of MAC addresses which it builds by learning what hosts are connected to each network. A
software bridge can be used within a Linux host in order to emulate a hardware bridge, for example
in virtualization applications for sharing a NIC with one or more virtual NICs. This case will be
illustrated here as an example.
To create a network bridge, create a file in the /et c/sysco n f ig /n et wo rk- scrip t s/ directory called
if cf g - b rN, replacing N with the number for the interface, such as 0.
The contents of the file is similar to whatever type of interface is getting bridged to, such as an
Ethernet interface. The differences in this example are as follows:
The D EVIC E directive is given an interface name as its argument in the format b rN, where N is
replaced with the number of the interface.

169

Deployment G uide

The T YPE directive is given an argument B rid g e. This directive determines the device type and
the argument is case sensitive.
The bridge interface configuration file now has the IP address and the physical interface has only
a MAC address.
An extra directive, D ELAY= 0, is added to prevent the bridge from waiting while it monitors traffic,
learns where hosts are located, and builds a table of MAC addresses on which to base its filtering
decisions. The default delay of 15 seconds is not needed if no routing loops are possible.
The N M_C O N T R O LLED = n o should be added to the Ethernet interface to prevent
N et wo rkMan ag er from altering the file. It can also be added to the bridge configuration file in
case future versions of N et wo rkMan ag er support bridge configuration.
The following is a sample bridge interface configuration file using a static IP address:

Examp le 10.4 . Samp le if cf g - b r0 in t erf ace co n f ig u rat io n f ile
D EVICE=br0
TYPE=Bridge
IPAD D R=192.168.1.1
NETMASK=255.255.255.0
ONBOOT=yes
BOOTPROTO=none
NM_CONTROLLED =no
D ELAY=0

To complete the bridge another interface is created, or an existing interface is modified, and pointed
to the bridge interface. The following is a sample Ethernet interface configuration file pointing to a
bridge interface. Configure your physical interface in /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h X,
where X is a unique number corresponding to a specific interface, as follows:

Examp le 10.5. Samp le if cf g - et h X in t erf ace co n f ig u rat io n f ile
D EVICE=ethX
TYPE=Ethernet
HWAD D R=AA:BB:CC:D D :EE:FF
BOOTPROTO=none
ONBOOT=yes
NM_CONTROLLED =no
BRID GE=br0

Note
For the D EVIC E directive, almost any interface name could be used as it does not determine
the device type. Other commonly used names include t ap , d u mmy and b o n d for example.
T YPE= Et h ern et is not strictly required. If the T YPE directive is not set, the device is treated
as an Ethernet device (unless its name explicitly matches a different interface configuration
file.)

170

⁠Chapt er 1 0 . Net work Int erfaces

You can see Section 10.2, “ Interface Configuration Files” for a review of the directives and options
used in network interface config files.

Warning
If you are configuring bridging on a remote host, and you are connected to that host over the
physical NIC you are configuring, please consider the implications of losing connectivity
before proceeding. You will lose connectivity when restarting the service and may not be able
to regain connectivity if any errors have been made. Console, or out-of-band access is
advised.
Restart the networking service, in order for the changes to take effect, as follows:
service n et wo rk rest art

1 0 .2 .6 .1 . Ne t wo rk Bridge wit h Bo nd
An example of a network bridge formed from two or more bonded Ethernet interfaces will now be given
as this is another common application in a virtualization environment. If you are not very familiar with
the configuration files for bonded interfaces then please see Section 10.2.4, “ Channel Bonding
Interfaces”
Create or edit two or more Ethernet interface configuration files, which are to be bonded, as follows:
D EVICE=ethX
TYPE=Ethernet
USERCTL=no
SLAVE=yes
MASTER=bond0
BOOTPROTO=none
HWAD D R=AA:BB:CC:D D :EE:FF
NM_CONTROLLED =no

Note
Using et h X as the interface name is common practice but almost any name could be used.
Names such as t ap , d u mmy and b o n d are commonly used.
Create or edit one interface configuration file, /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0, as
follows:
D EVICE=bond0
ONBOOT=yes
BOND ING_OPTS='mode=1 miimon=100'
BRID GE=br0
NM_CONTROLLED =no
For further instructions and advice on configuring the bonding module and to view the list of
bonding parameters, see Section 29.8.1, “ Using Channel Bonding” .
Create or edit one interface configuration file, /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b r0, as

171

Deployment G uide

follows:
D EVICE=br0
ONBOOT=yes
TYPE=Bridge
IPAD D R=192.168.1.1
NETMASK=255.255.255.0
NM_CONTROLLED =no

Fig u re 10.1. A n et wo rk b rid g e co n sist in g o f t wo b o n d ed Et h ern et in t erf aces.
We now have two or more interface configuration files with the MAST ER = b o n d 0 directive. These
point to the configuration file named /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b o n d 0, which
contains the D EVIC E= b o n d 0 directive. This if cf g - b o n d 0 in turn points to the
/et c/sysco n f ig /n et wo rk- scrip t s/if cf g - b r0 configuration file, which contains the IP address, and
acts as an interface to the virtual networks inside the host.
To bring up the new or recently configured interfaces, issue a command as ro o t in the following
format:
if u p device
Alternatively, restart the networking service, in order for the changes to take effect, as follows:

172

⁠Chapt er 1 0 . Net work Int erfaces

~]# service n et wo rk rest art

1 0 .2 .6 .2 . Ne t wo rk Bridge wit h Bo nde d VLAN
Virtualization servers that intend to have distinct subnets for its guests while still ensuring availability
in the event of a NIC failure will often combine bonds, VLANs, and bridges. An example of this
configuration will now be given. By creating a bridge on the VLAN instead of the underlying device
we allow VLAN tagging to be handled entirely through the host with no need to configure the guests'
interfaces.
1. Ensure the bond and VLAN have been configured as outlined in Section 10.2.5, “ Configuring
a VLAN over a Bond” .
2. Create the bridge's configuration file, if cf g - b r0:
~]# vi /etc/sysconfig/network-scripts/ifcfg-br0
D EVICE=br0
ONBOOT=yes
TYPE=Bridge
IPAD D R=192.168.10.1
NETMASK=255.255.255.0
NM_CONTROLLED =no
3. Adjust the VLAN's configuration file, if cf g - b o n d 0.19 2 from the earlier example, to use the
newly created b r0 as its master:
~]# vi /etc/sysconfig/network-scripts/ifcfg-bond0.192
D EVICE=bond0.192
BOOTPROTO=none
ONPARENT=yes
#IPAD D R=192.168.10.1
#NETMASK=255.255.255.0
VLAN=yes
NM_CONTROLLED =no
BRID GE=br0
4. To bring up the new or recently configured interfaces, issue a command as ro o t in the
following format:
if u p device
Alternatively, restart the networking service, in order for the changes to take effect, as follows:
~]# service n et wo rk rest art

10.2.7. Set t ing Up 802.1Q VLAN T agging
1. If required, start the VLAN 8021q module by issuing the following command as ro o t :
~]# mo d p ro b e - - f irst - t ime 8021q

173

Deployment G uide

No visual output indicates the module was not running and has now been loaded. Note that
given a correct configuration file, the VLAN 8021q module will be loaded as required and
therefore does not need to be loaded separately.
2. Configure your physical interface in /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h X, where X
is a unique number corresponding to a specific interface, as follows:
D EVICE=ethX
TYPE=Ethernet
BOOTPROTO=none
ONBOOT=yes
3. Configure the VLAN interface configuration in /et c/sysco n f ig /n et wo rk- scrip t s. The
configuration filename should be the physical interface plus a . character plus the VLAN ID
number. For example, if the VLAN ID is 192, and the physical interface is et h 0, then the
configuration filename should be if cf g - et h 0.19 2:
D EVICE=ethX.192
BOOTPROTO=none
ONBOOT=yes
IPAD D R=192.168.1.1
NETMASK=255.255.255.0
USERCTL=no
NETWORK=192.168.1.0
VLAN=yes
If there is a need to configure a second VLAN, with for example, VLAN ID 193, on the same
interface, et h 0 , add a new file with the name et h 0.19 3 with the VLAN configuration details.
4. Restart the networking service, in order for the changes to take effect. Issue the following
command as ro o t :
~]# service n et wo rk rest art

10.2.8. Alias and Clone Files
Two lesser-used types of interface configuration files are alias and clone files. As the ip utility now
supports assigning multiple addresses to the same interface it is no longer necessary to use this
method of binding multiple addresses to the same interface. The ip command to assign an address
can be repeated multiple times in order to assign multiple address. For example:
~]# ip ad d ress ad d 19 2.16 8.2.223/24 d ev et h 1
~]# ip ad d ress ad d 19 2.16 8.4 .223/24 d ev et h 1
~]# ip ad d r
3: eth1:  mtu 1500 qdisc pfifo_fast state UP qlen
1000
link/ether 52:54:00:fb:77:9e brd ff:ff:ff:ff:ff:ff
inet 192.168.2.223/24 scope global eth1
inet 192.168.4 .223/24 scope global eth1
The commands for the ip utility, sometimes referred to as iproute2 after the upstream package name,
are documented in the man ip ( 8) page. The package name in Red Hat Enterprise Linux 6 is iproute.

174

⁠Chapt er 1 0 . Net work Int erfaces

Note
In Red Hat Enterprise Linux 6, N et wo rkMan ag er now reads if cf g alias files and assigns the
addresses in them to their master interface, using the alias name as the address label. For
example, if if cf g - et h 0 and if cf g - et h 0:1 files are present, N et wo rkMan ag er reads the alias
file's D EVICE line and stores this as an address label. The use of secondary addresses rather
than alias is still preferred.
For new installations, users should select the Man u al method on the IPv4 or IPv6 tab in
N et wo rkMan ag er to assign multiple IP address to the same interface. For more information on
using this tool, see Chapter 9, NetworkManager.
Alias interface configuration files, which are used to bind multiple addresses to a single interface, use
the if cf g - if-name:alias-value naming scheme.
For example, an if cf g - et h 0:0 file could be configured to specify D EVIC E= et h 0:0 and a static IP
address of 10.0.0.2, serving as an alias of an Ethernet interface already configured to receive its IP
information via D H C P in if cf g - et h 0. Under this configuration, et h 0 is bound to a dynamic IP
address, but the same physical network card can receive requests via the fixed, 10.0.0.2 IP address.

Warning
Alias interfaces do not support D H C P.
A clone interface configuration file should use the following naming convention: if cf g - ifname- clone-name. While an alias file allows multiple addresses for an existing interface, a clone file
is used to specify additional options for an interface. For example, a standard D H C P Ethernet
interface called et h 0, may look similar to this:
D EVICE=eth0
ONBOOT=yes
BOOTPROTO=dhcp
Since the default value for the U SER C T L directive is n o if it is not specified, users cannot bring this
interface up and down. To give users the ability to control the interface, create a clone by copying
if cf g - et h 0 to if cf g - et h 0- u ser and add the following line to if cf g - et h 0- u ser:
USERCTL=yes
This way a user can bring up the et h 0 interface using the /sb in /if u p et h 0- u ser command because
the configuration options from if cf g - et h 0 and if cf g - et h 0- u ser are combined. While this is a very
basic example, this method can be used with a variety of options and interfaces.
It is no longer possible to create alias and clone interface configuration files using a graphical tool.
However, as explained at the beginning of this section, it is no longer necessary to use this method
as it is now possible to directly assign multiple IP address to the same interface. For new
installations, users should select the Man u al method on the IPv4 or IPv6 tab in N et wo rkMan ag er
to assign multiple IP address to the same interface. For more information on using this tool, see
Chapter 9, NetworkManager.

10.2.9. Dialup Int erfaces

175

Deployment G uide

If you are connecting to the Internet via a dialup connection, a configuration file is necessary for the
interface.
PPP interface files are named using the following format:
if cf g - p p p X
where X is a unique number corresponding to a specific interface.
The PPP interface configuration file is created automatically when wvd ial, or K p p p is used to create
a dialup account. It is also possible to create and edit this file manually.
The following is a typical /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - p p p 0 file:
D EVICE=ppp0
NAME=test
WVD IALSECT=test
MOD EMPORT=/dev/modem
LINESPEED =115200
PAPNAME=test
USERCTL=true
ONBOOT=no
PERSIST=no
D EFROUTE=yes
PEERD NS=yes
D EMAND =no
ID LETIMEOUT=600
Serial Line Internet Protocol (SLIP) is another dialup interface, although it is used less frequently. SLIP
files have interface configuration file names such as if cf g - sl0.
Other options that may be used in these files include:
D EFR O U T E= answer
where answer is one of the following:
yes — Set this interface as the default route.
n o — D o not set this interface as the default route.
D EMAN D = answer
where answer is one of the following:
yes — This interface allows p p p d to initiate a connection when someone attempts to
use it.
n o — A connection must be manually established for this interface.
ID LET IMEO U T = value
where value is the number of seconds of idle activity before the interface disconnects itself.
IN IT ST R IN G = string
where string is the initialization string passed to the modem device. This option is primarily
used in conjunction with SLIP interfaces.
LIN ESPEED = value

176

⁠Chapt er 1 0 . Net work Int erfaces

where value is the baud rate of the device. Possible standard values include 576 00, 384 00,
19 200, and 9 6 00.
MO D EMPO R T = device
where device is the name of the serial device that is used to establish the connection for the
interface.
MT U = value
where value is the Maximum Transfer Unit (MTU) setting for the interface. The MTU refers to
the largest number of bytes of data a frame can carry, not counting its header information.
In some dialup situations, setting this to a value of 576 results in fewer packets dropped
and a slight improvement to the throughput for a connection.
N AME= name
where name is the reference to the title given to a collection of dialup connection
configurations.
PAPN AME= name
where name is the user name given during the Password Authentication Protocol (PAP)
exchange that occurs to allow connections to a remote system.
PER SIST = answer
where answer is one of the following:
yes — This interface should be kept active at all times, even if deactivated after a modem
hang up.
n o — This interface should not be kept active at all times.
R EMIP= address
where address is the IP address of the remote system. This is usually left unspecified.
WVD IALSEC T = name
where name associates this interface with a dialer configuration in /et c/wvd ial.co n f . This
file contains the phone number to be dialed and other important information for the
interface.

10.2.10. Ot her Int erfaces
Other common interface configuration files include the following:
if cf g - lo
A local loopback interface is often used in testing, as well as being used in a variety of
applications that require an IP address pointing back to the same system. Any data sent to
the loopback device is immediately returned to the host's network layer.

177

Deployment G uide

Do not manually edit the ifcfg-lo script
The loopback interface script, /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - lo , should
never be edited manually. D oing so can prevent the system from operating correctly.
if cf g - irlan 0
An infrared interface allows information between devices, such as a laptop and a printer, to
flow over an infrared link. This works in a similar way to an Ethernet device except that it
commonly occurs over a peer-to-peer connection.
if cf g - p lip 0
A Parallel Line Interface Protocol (PLIP) connection works much the same way as an Ethernet
device, except that it utilizes a parallel port.
Interface configuration files for Linux on System z include the following:
if cf g - h siN
A HiperSockets interface is an interface for high-speed TCP/IP communication within and
across z/VM guest virtual machines and logical partitions (LPARs) on an IBM System z
mainframe.

10.3. Int erface Cont rol Script s
The interface control scripts activate and deactivate system interfaces. There are two primary
interface control scripts that call on control scripts located in the /et c/sysco n f ig /n et wo rk- scrip t s/
directory: /sb in /if d o wn and /sb in /if u p .
The if u p and if d o wn interface scripts are symbolic links to scripts in the /sb in / directory. When
either of these scripts are called, they require the value of the interface to be specified, such as:
if u p et h 0

Use the ifup and ifdown interface scripts
The if u p and if d o wn interface scripts are the only scripts that the user should use to bring
up and take down network interfaces.
The following scripts are described for reference purposes only.

Two files used to perform a variety of network initialization tasks during the process of bringing up a
network interface are /et c/rc.d /in it .d /f u n ct io n s and /et c/sysco n f ig /n et wo rk- scrip t s/n et wo rkf u n ct io n s. See Section 10.7, “ Network Function Files” for more information.
After verifying that an interface has been specified and that the user executing the request is allowed
to control the interface, the correct script brings the interface up or down. The following are common
interface control scripts found within the /et c/sysco n f ig /n et wo rk- scrip t s/ directory:
if u p - aliases
Configures IP aliases from interface configuration files when more than one IP address is

178

⁠Chapt er 1 0 . Net work Int erfaces

Configures IP aliases from interface configuration files when more than one IP address is
associated with an interface.
if u p - ip p p an d if d o wn - ip p p
Brings ISD N interfaces up and down.
if u p - ip v6 an d if d o wn - ip v6
Brings IPv6 interfaces up and down.
if u p - p lip
Brings up a PLIP interface.
if u p - p lu sb
Brings up a USB interface for network connections.
if u p - p o st an d if d o wn - p o st
Contains commands to be executed after an interface is brought up or down.
if u p - p p p an d if d o wn - p p p
Brings a PPP interface up or down.
if u p - ro u t es
Adds static routes for a device as its interface is brought up.
if d o wn - sit an d if u p - sit
Contains function calls related to bringing up and down an IPv6 tunnel within an IPv4
connection.
if u p - wireless
Brings up a wireless interface.

Be careful when removing or modifying network scripts!
Removing or modifying any scripts in the /et c/sysco n f ig /n et wo rk- scrip t s/ directory can
cause interface connections to act irregularly or fail. Only advanced users should modify
scripts related to a network interface.
The easiest way to manipulate all network scripts simultaneously is to use the /sb in /service
command on the network service (/et c/rc.d /in it .d /n et wo rk), as illustrated by the following
command:
/sb in /service n et wo rk action
Here, action can be either st art , st o p , or rest art .
To view a list of configured devices and currently active network interfaces, use the following
command:
/sb in /service n et wo rk st at u s

179

Deployment G uide

10.4 . St at ic Rout es and t he Default Gat eway
Static routes are for traffic that must not, or should not, go through the default gateway. Routing is
often handled by devices on the network dedicated to routing (although any device can be
configured to perform routing). Therefore, it is often not necessary to configure static routes on
Red Hat Enterprise Linux servers or clients. Exceptions include traffic that must pass through an
encrypted VPN tunnel or traffic that should take a specific route for reasons of cost or security. The
default gateway is for any and all traffic which is not destined for the local network and for which no
preferred route is specified in the routing table. The default gateway is traditionally a dedicated
network router.

Configuring St at ic Rout es Using t he Command Line
If static routes are required, they can be added to the routing table by means of the ip ro u t e ad d
command and removed using the ip ro u t e d el command. The more frequently used ip ro u t e
commands take the following form:
ip route [ add | del | change | append | replace ] destination-address
See the ip - ro u t e( 8) man page for more details on the options and formats.
Use the ip ro u t e command without options to display the IP routing table. For example:
~]$ ip route
default via 192.168.122.1 dev eth0 proto static metric 1024
192.168.122.0/24 dev ens9 proto kernel scope link src 192.168.122.107
192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.126
To add a static route to a host address, in other words to a single IP address, issue a command as
ro o t :
~]# ip route add 192.0.2.1 via 10.0.0.1 [d ev ifname]
Where 192.0.2.1 is the IP address of the host in dotted decimal notation, 10.0.0.1 is the next hop
address and ifname is the exit interface leading to the next hop.
To add a static route to a network, in other words to an IP address representing a range of IP
addresses, issue the following command as ro o t :
~]# ip route add 192.0.2.0/24 via 10.0.0.1 [d ev ifname]
where 192.0.2.0 is the IP address of the destination network in dotted decimal notation and /24 is the
network prefix. The network prefix is the number of enabled bits in the subnet mask. This format of
network address slash network prefix length is sometimes referred to as classless inter-domain routing
(CID R) notation.
Static route configuration can be stored per-interface in a /et c/sysco n f ig /n et wo rkscrip t s/ro u t e- interface file. For example, static routes for the eth0 interface would be stored in the
/et c/sysco n f ig /n et wo rk- scrip t s/ro u t e- et h 0 file. The ro u t e- interface file has two formats: ip
command arguments and network/netmask directives. These are described below.
See the ip - ro u t e( 8) man page for more information on the ip ro u t e command.

Configuring T he Default Gat eway

180

⁠Chapt er 1 0 . Net work Int erfaces

The default gateway is determined by the network scripts which parse the /et c/sysco n f ig /n et wo rk
file first and then the network interface if cf g files for interfaces that are “ up” . The if cf g files are
parsed in numerically ascending order, and the last GATEWAY directive to be read is used to
compose a default route in the routing table.
The default route can thus be indicated by means of the GATEWAY directive and can be specified
either globally or in interface-specific configuration files. Specifying the gateway globally has certain
advantages in static networking environments, especially if more than one network interface is
present. It can make fault finding simpler if applied consistently. There is also the GATEWAYD EV
directive, which is a global option. If multiple devices specify GATEWAY, and one interface uses the
GATEWAYD EV directive, that directive will take precedence. This option is not recommend as it can
have unexpected consequences if an interface goes down and it can complicate fault finding.
In dynamic network environments, where mobile hosts are managed by N et wo rkMan ag er, gateway
information is likely to be interface specific and is best left to be assigned by D H C P. In special cases
where it is necessary to influence N et wo rkMan ag er's selection of the exit interface to be used to
reach a gateway, make use of the D EFR O U T E= n o command in the if cf g files for those interfaces
which do not lead to the default gateway.
Global default gateway configuration is stored in the /et c/sysco n f ig /n et wo rk file. This file specifies
gateway and host information for all network interfaces. For more information about this file and the
directives it accepts, see Section D .1.13, “ /etc/sysconfig/network” .

10.5. Configuring St at ic Rout es in ifcfg files
Static routes set using ip commands at the command prompt will be lost if the system is shutdown or
restarted. To configure static routes to be persistent after a system restart, they must be placed in perinterface configuration files in the /et c/sysco n f ig /n et wo rk- scrip t s/ directory. The file name should
be of the format ro u t e- ifname. There are two types of commands to use in the configuration files; ip
commands as explained in Section 10.5.1, “ Static Routes Using the IP Command Arguments Format”
and the Network/Netmask format as explained in Section 10.5.2, “ Network/Netmask D irectives
Format” .

10.5.1. St at ic Rout es Using t he IP Command Argument s Format
If required in a per-interface configuration file, for example /et c/sysco n f ig /n et wo rkscrip t s/ro u t e- et h 0, define a route to a default gateway on the first line. This is only required if the
gateway is not set via D H C P and is not set globally in the /et c/sysco n f ig /n et wo rk file:
default via 192.168.1.1 d ev interface
where 192.168.1.1 is the IP address of the default gateway. The interface is the interface that is
connected to, or can reach, the default gateway. The d ev option can be omitted, it is optional. Note
that this setting takes precedence over a setting in the /et c/sysco n f ig /n et wo rk file.
If a route to a remote network is required, a static route can be specified as follows. Each line is
parsed as an individual route:
10.10.10.0/24 via 192.168.1.1 [d ev interface]
where 10.10.10.0/24 is the network address and prefix length of the remote or destination network. The
address 192.168.1.1 is the IP address leading to the remote network. It is preferably the next hop
address but the address of the exit interface will work. The “ next hop” means the remote end of a link,
for example a gateway or router. The d ev option can be used to specify the exit interface interface but
it is not required. Add as many static routes as required.

181

Deployment G uide

The following is an example of a ro u t e- interface file using the ip command arguments format. The
default gateway is 19 2.16 8.0.1, interface eth0 and a leased line or WAN connection is available at
19 2.16 8.0.10. The two static routes are for reaching the 10.10.10.0/24 network and the
172.16 .1.10/32 host:
default via 192.168.0.1 dev eth0
10.10.10.0/24 via 192.168.0.10 dev eth0
172.16.1.10/32 via 192.168.0.10 dev eth0
In the above example, packets going to the local 19 2.16 8.0.0/24 network will be directed out the
interface attached to that network. Packets going to the 10.10.10.0/24 network and 172.16 .1.10/32
host will be directed to 19 2.16 8.0.10. Packets to unknown, remote, networks will use the default
gateway therefore static routes should only be configured for remote networks or hosts if the default
route is not suitable. Remote in this context means any networks or hosts that are not directly
attached to the system.
Specifying an exit interface is optional. It can be useful if you want to force traffic out of a specific
interface. For example, in the case of a VPN, you can force traffic to a remote network to pass through
a tun0 interface even when the interface is in a different subnet to the destination network.

Duplicate default gateways
If the default gateway is already assigned from D H C P, the IP command arguments format can
cause one of two errors during start-up, or when bringing up an interface from the down state
using the if u p command: " RTNETLINK answers: File exists" or 'Error: either " to" is a duplicate,
or " X.X.X.X" is a garbage.', where X.X.X.X is the gateway, or a different IP address. These
errors can also occur if you have another route to another network using the default gateway.
Both of these errors are safe to ignore.

10.5.2. Net work/Net mask Direct ives Format
You can also use the network/netmask directives format for ro u t e- interface files. The following is a
template for the network/netmask format, with instructions following afterwards:
AD D RESS0=10.10.10.0
NETMASK0=255.255.255.0
GATEWAY0=192.168.1.1
AD D R ESS0= 10.10.10.0 is the network address of the remote network or host to be reached.
N ET MASK 0= 255.255.255.0 is the netmask for the network address defined with
AD D R ESS0= 10.10.10.0.
G AT EWAY0= 192.168.1.1 is the default gateway, or an IP address that can be used to reach
AD D R ESS0= 10.10.10.0
The following is an example of a ro u t e- interface file using the network/netmask directives format.
The default gateway is 19 2.16 8.0.1 but a leased line or WAN connection is available at
19 2.16 8.0.10. The two static routes are for reaching the 10.10.10.0/24 and 172.16 .1.0/24
networks:
AD D RESS0=10.10.10.0
NETMASK0=255.255.255.0

182

⁠Chapt er 1 0 . Net work Int erfaces

GATEWAY0=192.168.0.10
AD D RESS1=172.16.1.10
NETMASK1=255.255.255.0
GATEWAY1=192.168.0.10
Subsequent static routes must be numbered sequentially, and must not skip any values. For
example, AD D R ESS0, AD D R ESS1, AD D R ESS2, and so on.

10.6. Configuring IPv6 T okeniz ed Int erface Ident ifiers
In a network, servers are generally given static addresses and these are usually configured manually
to avoid relying on a D H C P server which may fail or run out of addresses. The IPv6 protocol
introduced Stateless Address Autoconfiguration (SLAAC) which enables clients to assign themselves an
address without relying on a D H C Pv6 server. SLAAC derives the IPv6 address based on the
interface hardware, therefore it should not be used for servers in case the hardware is changed and
the associated SLAAC generated address changes with it. In an IPv6 environment, if the network
prefix is changed, or the system is moved to a new location, any manually configured static
addresses would have to be edited due to the changed prefix.
To address these problems, the IETF draft Tokenised IPv6 Identifiers has been implemented in the
kernel together with corresponding additions to the ip utility. This enables the lower 64 bit interface
identifier part of the IPv6 address to be based on a token, supplied by the administrator, leaving the
network prefix, the higher 64 bits, to be obtained from router advertisements (RA). This means that if
the network interface hardware is changed, the lower 64 bits of the address will not change, and if the
system is moved to another network, the network prefix will be obtained from router advertisements
automatically, thus no manual editing is required.
To configure an interface to use a tokenized IPv6 identifier, issue a command in the following format
as ro o t user:
~]# ip t o ken set ::1a:2b :3c:4 d /6 4 d ev et h 4
Where ::1a:2b :3c:4 d /6 4 is the token to be used. This setting is not persistent. To make it persistent,
add the command to an init script. See Section 10.3, “ Interface Control Scripts” .
Using a memorable token is possible, but is limited to the range of valid hexadecimal digits. For
example, for a D N S server, which traditionally uses port 53, a token of ::53/6 4 could be used.
To view all the configured IPv6 tokens, issue the following command:
~]$ ip t o ken
token :: dev eth0
token :: dev eth1
token :: dev eth2
token :: dev eth3
token ::1a:2b:3c:4d dev eth4
To view the configured IPv6 token for a specific interface, issue the following command:
~]$ ip t o ken g et d ev et h 4
token ::1a:2b:3c:4d dev eth4

183

Deployment G uide

Note that adding a token to an interface will replace a previously allocated token, and in turn
invalidate the address derived from it. Supplying a new token causes a new address to be generated
and applied, but this process will leave any other addresses unchanged. In other words, a new
tokenized identifier only replaces a previously existing tokenized identifier, not any other IP address.

Note
Take care not to add the same token to more than one system or interface as the duplicate
address detection (D AD ) mechanism will not be able to resolve the problem. Once a token is
set, it cannot be cleared or reset, except by rebooting the machine.

10.7. Net work Funct ion Files
Red Hat Enterprise Linux makes use of several files that contain important common functions used to
bring interfaces up and down. Rather than forcing each interface control file to contain these
functions, they are grouped together in a few files that are called upon when necessary.
The /et c/sysco n f ig /n et wo rk- scrip t s/n et wo rk- f u n ct io n s file contains the most commonly used
IPv4 functions, which are useful to many interface control scripts. These functions include
contacting running programs that have requested information about changes in the status of an
interface, setting host names, finding a gateway device, verifying whether or not a particular device is
down, and adding a default route.
As the functions required for IPv6 interfaces are different from IPv4 interfaces, a
/et c/sysco n f ig /n et wo rk- scrip t s/n et wo rk- f u n ct io n s- ip v6 file exists specifically to hold this
information. The functions in this file configure and delete static IPv6 routes, create and remove
tunnels, add and remove IPv6 addresses to an interface, and test for the existence of an IPv6
address on an interface.

10.8. Et ht ool
Et h t o o l is a utility for configuration of Network Interface Cards (NICs). This utility allows querying
and changing settings such as speed, port, auto-negotiation, PCI locations and checksum offload
on many network devices, especially Ethernet devices.
We present here a short selection of often used et h t o o l commands together with some useful
commands that are not well known. For a full list of commands type et h t o o l - h or see the man page,
et h t o o l( 8) , for a more comprehensive list and explanation. The first two examples are information
queries and show the use of the different formats of the command.
But first, the command structure:
et h t o o l [option...] devname
where option is none or more options, and devname is your Network Interface Card (NIC). For example
eth0 or em1.
et h t o o l
The et h t o o l command with only a device name as an option is used to print the current
settings of the specified device. It takes the following form:
ethtool devname

184

⁠Chapt er 1 0 . Net work Int erfaces

where devname is your NIC. For example eth0 or em1.
Some values can only be obtained when the command is run as ro o t . Here is an example of the
output when the command is run as ro o t :
~]# et h t o o l em1
Settings for em1:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Speed: 1000Mb/s
D uplex: Full
Port: Twisted Pair
PHYAD : 2
Transceiver: internal
Auto-negotiation: on
MD I-X: on
Supports Wake-on: pumbg
Wake-on: g
Current message level: 0x00000007 (7)
drv probe link
Link detected: yes
Issue the following command, using the short or long form of the argument, to query the specified
network device for associated driver information:
et h t o o l - i, - - d river devname
where devname is your Network Interface Card (NIC). For example eth0 or em1.
Here is an example of the output:
~]$ et h t o o l - i em1
driver: e1000e
version: 2.0.0-k
firmware-version: 0.13-3
bus-info: 0000:00:19.0
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
Here follows a list of command options to query, identify or reset the device. They are in the usual sh o rt and - - lo n g form:
- - st at ist ics
The - - st at ist ics or - S queries the specified network device for NIC and driver statistics. It

185

Deployment G uide

The - - st at ist ics or - S queries the specified network device for NIC and driver statistics. It
takes the following form:
- S, - - st at ist ics devname
where devname is your NIC.
- - id en t if y
The - - id en t if y or - p option initiates adapter-specific action intended to enable an operator
to easily identify the adapter by sight. Typically this involves blinking one or more LED s on
the specified network port. It takes the following form:
- p , - - id en t if y devname integer
where integer is length of time in seconds to perform the action,
and devname is your NIC.
- - sh o w- t ime- st amp in g
The - - sh o w- t ime- st amp in g or - T option queries the specified network device for time
stamping parameters. It takes the following form:
- T , - - sh o w- t ime- st amp in g devname
where devname is your NIC.
- - sh o w- o f f lo ad
The - - sh o w- f eat u res, or - - sh o w- o f f lo ad , or - k option queries the specified network
device for the state of protocol offload and other features. It takes the following form:
- k, - - sh o w- f eat u res, - - sh o w- o f f lo ad devname
where devname is your NIC.
- - t est
The - - t est or - t option is used to perform tests on a Network Interface Card. It takes the
following form:
- t , - - t est devname word
where word is one of the following:
o f f lin e — Perform a comprehensive set of tests. Service will be interrupted.
o n lin e — Perform a reduced set of tests. Service should not be interrupted.
ext ern al_lb — Perform full set of tests including loopback tests while fitted with a
loopback cable.
and devname is your NIC.

186

⁠Chapt er 1 0 . Net work Int erfaces

Changing some or all settings of the specified network device requires the - s or - - ch an g e option. All
the following options are only applied if the - s or - - ch an g e option is also specified. For the sake of
clarity we will omit it here.
To make these settings permanent you can make use of the ET H T O O L_O PT S directive. It can be
used in interface configuration files to set the desired options when the network interface is brought
up. See Section 10.2.1, “ Ethernet Interfaces” for more details on how to use this directive.
- - o f f lo ad
The - - f eat u res, or - - o f f lo ad , or - K option changes the offload parameters and other
features of the specified network device. It takes the following form:
- K , - - f eat u res, - - o f f lo ad devname feature boolean
where feature is a built-in or kernel supplied feature,
boolean is one of O N or O FF,
and devname is your NIC.
The et h t o o l( 8) man page lists most features. As the feature set is dependent on the NIC
driver, you should consult the driver documentation for features not listed in the man page.
- - sp eed
The - - sp eed option is used to set the speed in megabits per second (Mb/s). Omitting the
speed value will show the supported device speeds. It takes the following form:
- - sp eed number devname
where number is the speed in megabits per second (Mb/s),
and devname is your NIC.
- - d u p lex
The - - d u p lex option is used to set the transmit and receive mode of operation. It takes the
following form:
- - d u p lex word devname
where word is one of the following:
h alf — Sets half-duplex mode. Usually used when connected to a hub.
f u ll — Sets full-duplex mode. Usually used when connected to a switch or another host.
and devname is your NIC.
- - p o rt
The - - p o rt option is used to select the device port . It takes the following form:
- - p o rt value devname
where value is one of the following:

187

Deployment G uide

t p — An Ethernet interface using Twisted-Pair cable as the medium.
au i — Attachment Unit Interface (AUI). Normally used with hubs.
b n c — An Ethernet interface using BNC connectors and co-axial cable.
mii — An Ethernet interface using a Media Independent Interface (MII).
f ib re — An Ethernet interface using Optical Fibre as the medium.
and devname is your NIC.
- - au t o n eg
The - - au t o n eg option is used to control auto-negotiation of network speed and mode of
operation (full-duplex or half-duplex mode). If auto-negotiation is enabled you can initiate
re-negotiation of network speeds and mode of operation by using the - r, - - n eg o t iat e
option. You can display the auto-negotiation state using the - - a, - - sh o w- p au se option.
It takes the following form:
- - au t o n eg value devname
where value is one of the following:
yes — Allow auto-negotiating of network speed and mode of operation.
n o — D o not allow auto-negotiating of network speed and mode of operation.
and devname is your NIC.
- - ad vert ise
The - - ad vert ise option is used to set what speeds and modes of operation (duplex mode)
are advertised for auto-negotiation. The argument is one or more hexadecimal values from
Table 10.1, “ Ethtool advertise options: speed and mode of operation” .
It takes the following form:
- - ad vert ise option devname
where option is one or more of the hexadecimal values from the table below and devname is
your NIC.
T ab le 10.1. Et h t o o l ad vert ise o p t io n s: sp eed an d mo d e o f o p erat io n

188

H ex Valu e

Sp eed

D u p lex Mo d e

IEEE st an d ard ?

0x001
0x002
0x004
0x008
0x010
0x020
0x8000
0x1000
0x20000
0x20000

10
10
100
100
1000
1000
2500
10000
20000MLD 2
20000MLD 2

Half
Full
Half
Full
Half
Full
Full
Full
Full
Full

Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
No
No

⁠Chapt er 1 0 . Net work Int erfaces

H ex Valu e
0x40000
- - p h yad

Sp eed

D u p lex Mo d e

IEEE st an d ard ?

20000KR2

Full

No

The - - p h yad option is used to change the physical address. Often referred to as the MAC
or hardware address but in this context referred to as the physical address.
It takes the following form:
- - p h yad physical_address devname
where physical_address is the physical address in hexadecimal format and devname is your
NIC.
- - xcvr
The - - xcvr option is used to select the transceiver type. Currently only “ internal” and
“ external” can be specified. In the future other types might be added.
It takes the following form:
- - xcvr word devname
where word is one of the following:
in t ern al — Use internal transceiver.
ext ern al — Use external transceiver.
and devname is your NIC.
- - wo l
The - - wo l option is used to set “ Wake-on-LAN” options. Not all devices support this. The
argument to this option is a string of characters specifying which options to enable.
It takes the following form:
- - wo l value devname
where value is one or more of the following:
p — Wake on PHY activity.
u — Wake on unicast messages.
m — Wake on multicast messages.
b — Wake on broadcast messages.
g — Wake-on-Lan; wake on receipt of a " magic packet" .
s — Enable security function using password for Wake-on-Lan.
d — D isable Wake-on-Lan and clear all settings.
and devname is your NIC.

189

Deployment G uide

- - so p ass
The - - so p ass option is used to set the “ SecureOn” password. The argument to this option
must be 6 bytes in Ethernet MAC hexadecimal format (xx:yy:zz:aa:bb:cc).
It takes the following form:
- - so p ass xx:yy:zz:aa:bb:cc devname
where xx:yy:zz:aa:bb:cc is the password in the same format as a MAC address and devname
is your NIC.
- - msg lvl
The - - msg lvl option is used to set the driver message-type flags by name or number. The
precise meanings of these type flags differ between drivers.
It takes the following form:
- - msg lvl message_type devname
where message_type is one of:
message type name in plain text.
hexadecimal number indicating the message type.
and devname is your NIC.
The defined message type names and numbers are shown in the table below:
T ab le 10.2. D river messag e t yp e
Messag e T yp e

H ex Valu e

D escrip t io n

drv
probe
link
timer
ifdown

0x0001
0x0002
0x0004
0x0008
0x0010

ifup
rx_err
tx_err
intr
tx_done
rx_status
pktdata
hw
wol

0x0020
0x0040
0x0080
0x0200
0x0400
0x0800
0x1000
0x2000
0x4000

General driver status
Hardware probing
Link state
Periodic status check
Interface being brought
down
Interface being brought up
Receive error
Transmit error
Interrupt handling
Transmit completion
Receive completion
Packet contents
Hardware status
Wake-on-LAN status

10.9. Addit ional Resources
The following are resources which explain more about network interfaces.

190

⁠Chapt er 1 0 . Net work Int erfaces

Inst alled Document at ion
/u sr/sh are/d o c/in it scrip t s- version/sysco n f ig .t xt — A guide to available options for network
configuration files, including IPv6 options not covered in this chapter.

Online Resources
http://linux-ip.net/gl/ip-cref/ — This document contains a wealth of information about the ip
command, which can be used to manipulate routing tables, among other things.
Red Hat Access Labs — The Red Hat Access Labs includes a “ Network Bonding Helper” .

See Also
Appendix E, The proc File System — D escribes the sysct l utility and the virtual files within the
/p ro c/ directory, which contain networking parameters and statistics among other things.

191

Deployment G uide

⁠Part V. Infrastructure Services
This part provides information how to configure services and daemons, configure authentication,
and enable remote logins.

192

⁠Chapt er 1 1 . Services and Daemons

Chapter 11. Services and Daemons
Maintaining security on your system is extremely important, and one approach for this task is to
manage access to system services carefully. Your system may need to provide open access to
particular services (for example, h t t p d if you are running a web server). However, if you do not need
to provide a service, you should turn it off to minimize your exposure to possible bug exploits.
This chapter explains the concept of runlevels, and describes how to set the default one. It also
covers the setup of the services to be run in each of these runlevels, and provides information on
how to start, stop, and restart the services on the command line using the service command.

Keep the system secure
When you allow access for new services, always remember that both the firewall and SELin u x
need to be configured as well. One of the most common mistakes committed when configuring
a new service is neglecting to implement the necessary firewall configuration and SELinux
policies to allow access for it. For more information, see the Red Hat Enterprise Linux 6 Security
Guide.

11.1. Configuring t he Default Runlevel
A runlevel is a state, or mode, defined by services that are meant to be run when this runlevel is
selected. Seven numbered runlevels exist (indexed from 0):
T ab le 11.1. R u n levels in R ed H at En t erp rise Lin u x
R u n level

D escrip t io n

0
1
2
3
4
5
6

Used to halt the system. This runlevel is reserved and cannot be changed.
Used to run in a single-user mode. This runlevel is reserved and cannot be changed.
Not used by default. You are free to define it yourself.
Used to run in a full multi-user mode with a command-line user interface.
Not used by default. You are free to define it yourself.
Used to run in a full multi-user mode with a graphical user interface.
Used to reboot the system. This runlevel is reserved and cannot be changed.

To check in which runlevel you are operating, type the following:
~]$ ru n level
N5
The ru n level command displays previous and current runlevel. In this case it is number 5, which
means the system is running in a full multi-user mode with a graphical user interface.
The default runlevel can be changed by modifying the /et c/in it t ab file, which contains a line near
the end of the file similar to the following:
id:5:initdefault:
To do so, edit this file as ro o t and change the number on this line to the desired value. The change
will take effect the next time you reboot the system.

193

Deployment G uide

11.2. Configuring t he Services
To allow you to configure which services are started at boot time, Red Hat Enterprise Linux is shipped
with the following utilities: the Service C o n f ig u rat io n graphical application, the n t sysv text user
interface, and the ch kco n f ig command-line tool.

Enabling the irqbalance service
To ensure optimal performance on POWER architecture, it is recommended that the
irq b alan ce service is enabled. In most cases, this service is installed and configured to run
during the Red Hat Enterprise Linux 6 installation. To verify that irq b alan ce is running, as
ro o t , type the following at a shell prompt:
~]# service irq b alan ce st at u s
irqbalance (pid 1234) is running...
For information on how to enable and run a service using a graphical user interface, see
Section 11.2.1, “ Using the Service Configuration Utility” . For instructions on how to perform
these task on the command line, see Section 11.2.3, “ Using the chkconfig Utility” and
Section 11.3, “ Running Services” respectively.

11.2.1. Using t he Service Configurat ion Ut ilit y
The Service C o n f ig u rat io n utility is a graphical application developed by Red Hat to configure
which services are started in a particular runlevel, as well as to start, stop, and restart them from the
menu. To start the utility, select Syst em → Ad min ist rat io n → Services from the panel, or type the
command syst em- co n f ig - services at a shell prompt.

Note
The syst em- co n f ig - services utility is provided by the system-config-services package,
which may not be installed by default on your version of Red Hat Enterprise Linux. To ensure
that, first run the following command:
~]$ rp m - q syst em- co n f ig - services
If the package is not installed by default, install it manually by running the following command
as root:
~]# yu m in st all syst em- co n f ig - services

194

⁠Chapt er 1 1 . Services and Daemons

Fig u re 11.1. T h e Service C o n f ig u rat io n u t ilit y
The utility displays the list of all available services (services from the /et c/rc.d /in it .d / directory, as
well as services controlled by xin et d ) along with their description and the current status. For a
complete list of used icons and an explanation of their meaning, see Table 11.2, “ Possible service
states” .
Note that unless you are already authenticated, you will be prompted to enter the superuser
password the first time you make a change.
T ab le 11.2. Po ssib le service st at es
Ico n

D escrip t io n
The service is enabled.
The service is disabled.
The service is enabled for selected runlevels only.
The service is running.
The service is stopped.
There is something wrong with the service.
The status of the service is unknown.

1 1 .2 .1 .1 . Enabling and Disabling a Se rvice
To enable a service, select it from the list and either click the En ab le button on the toolbar, or choose
Service → En ab le from the main menu.

195

Deployment G uide

To disable a service, select it from the list and either click the D isab le button on the toolbar, or
choose Service → D isab le from the main menu.

1 1 .2 .1 .2 . St art ing, Re st art ing, and St o pping a Se rvice
To start a service, select it from the list and either click the St art button on the toolbar, or choose
Service → St art from the main menu. Note that this option is not available for services controlled by
xin et d , as they are started by it on demand.
To restart a running service, select it from the list and either click the R est art button on the toolbar,
or choose Service → R est art from the main menu. Note that this option is not available for services
controlled by xin et d , as they are started and stopped by it automatically.
To stop a service, select it from the list and either click the St o p button on the toolbar, or choose
Service → St o p from the main menu. Note that this option is not available for services controlled by
xin et d , as they are stopped by it when their job is finished.

1 1 .2 .1 .3. Se le ct ing Runle ve ls
To enable the service for certain runlevels only, select it from the list and either click the C u st o miz e
button on the toolbar, or choose Service → C u st o miz e from the main menu. Then select the
checkbox beside each runlevel in which you want the service to run. Note that this option is not
available for services controlled by xin et d .

11.2.2. Using t he nt sysv Ut ilit y
The n t sysv utility is a command-line application with a simple text user interface to configure which
services are to be started in selected runlevels. To start the utility, type n t sysv at a shell prompt as
ro o t .

Fig u re 11.2. T h e n t sysv u t ilit y

196

⁠Chapt er 1 1 . Services and Daemons

The utility displays the list of available services (the services from the /et c/rc.d /in it .d / directory)
along with their current status and a description obtainable by pressing F1. For a list of used
symbols and an explanation of their meaning, see Table 11.3, “ Possible service states” .
T ab le 11.3. Po ssib le service st at es
Symb o l

D escrip t io n

[ *]
[]

The service is enabled.
The service is disabled.

1 1 .2 .2 .1 . Enabling and Disabling a Se rvice
To enable a service, navigate through the list using the U p and D o wn arrows keys, and select it with
the Sp aceb ar. An asterisk (*) appears in the brackets.
To disable a service, navigate through the list using the U p and D o wn arrows keys, and toggle its
status with the Sp aceb ar. An asterisk (*) in the brackets disappears.
Once you are done, use the T ab key to navigate to the O k button, and confirm the changes by
pressing En t er. Keep in mind that n t sysv does not actually start or stop the service. If you need to
start or stop the service immediately, use the service command as described in Section 11.3.2,
“ Starting a Service” .

1 1 .2 .2 .2 . Se le ct ing Runle ve ls
By default, the n t sysv utility only affects the current runlevel. To enable or disable services for other
runlevels, as ro o t , run the command with the additional - - level option followed by numbers from 0
to 6 representing each runlevel you want to configure:
n t sysv - - level runlevels
For example, to configure runlevels 3 and 5, type:
~]# n t sysv - - level 35

11.2.3. Using t he chkconfig Ut ilit y
The ch kco n f ig utility is a command-line tool that allows you to specify in which runlevel to start a
selected service, as well as to list all available services along with their current setting. Note that with
the exception of listing, you must have superuser privileges to use this command.

1 1 .2 .3.1 . List ing t he Se rvice s
To display a list of system services (services from the /et c/rc.d /in it .d / directory, as well as the
services controlled by xin et d ), either type ch kco n f ig - - list , or use ch kco n f ig with no additional
arguments. You will be presented with an output similar to the following:
~]# ch kco n f ig - - list
NetworkManager 0:off 1:off 2:on 3:on 4:on 5:on 6:off
abrtd
0:off 1:off 2:off 3:on 4:off 5:on 6:off
acpid
0:off 1:off 2:on 3:on 4:on 5:on 6:off
anamon
0:off 1:off 2:off 3:off 4:off 5:off 6:off
atd
0:off 1:off 2:off 3:on 4:on 5:on 6:off
auditd
0:off 1:off 2:on 3:on 4:on 5:on 6:off

197

Deployment G uide

avahi-daemon 0:off 1:off 2:off 3:on 4:on 5:on 6:off
... several lines omitted ...
wpa_supplicant 0:off 1:off 2:off 3:off 4:off 5:off 6:off
xinetd based services:
chargen-dgram: off
chargen-stream: off
cvs:
off
daytime-dgram: off
daytime-stream: off
discard-dgram: off
... several lines omitted ...
time-stream: off
Each line consists of the name of the service followed by its status (on or off) for each of the seven
numbered runlevels. For example, in the listing above, N et wo rkMan ag er is enabled in runlevel 2, 3,
4, and 5, while ab rt d runs in runlevel 3 and 5. The xin et d based services are listed at the end,
being either on, or off.
To display the current settings for a selected service only, use ch kco n f ig - - list followed by the
name of the service:
ch kco n f ig - - list service_name
For example, to display the current settings for the ssh d service, type:
~]# ch kco n f ig - - list ssh d
sshd
0:off 1:off 2:on

3:on

4:on

5:on

6:off

You can also use this command to display the status of a service that is managed by xin et d . In that
case, the output will only contain the information whether the service is enabled or disabled:
~]# ch kco n f ig - - list rsyn c
rsync
off

1 1 .2 .3.2 . Enabling a Se rvice
To enable a service in runlevels 2, 3, 4, and 5, type the following at a shell prompt as ro o t :
ch kco n f ig service_name o n
For example, to enable the h t t p d service in these four runlevels, type:
~]# ch kco n f ig h t t p d o n
To enable a service in certain runlevels only, add the - - level option followed by numbers from 0 to 6
representing each runlevel in which you want the service to run:
ch kco n f ig service_name o n - - level runlevels
For instance, to enable the ab rt d service in runlevels 3 and 5, type:
~]# ch kco n f ig ab rt d o n - - level 35

198

⁠Chapt er 1 1 . Services and Daemons

The service will be started the next time you enter one of these runlevels. If you need to start the
service immediately, use the service command as described in Section 11.3.2, “ Starting a Service” .
D o not use the - - level option when working with a service that is managed by xin et d , as it is not
supported. For example, to enable the rsyn c service, type:
~]# ch kco n f ig rsyn c o n
If the xin et d daemon is running, the service is immediately enabled without having to manually
restart the daemon.

1 1 .2 .3.3. Disabling a Se rvice
To disable a service in runlevels 2, 3, 4, and 5, type the following at a shell prompt as ro o t :
ch kco n f ig service_name o f f
For instance, to disable the h t t p d service in these four runlevels, type:
~]# ch kco n f ig h t t p d o f f
To disable a service in certain runlevels only, add the - - level option followed by numbers from 0 to 6
representing each runlevel in which you do not want the service to run:
ch kco n f ig service_name o f f - - level runlevels
For instance, to disable the ab rt d in runlevels 2 and 4, type:
~]# ch kco n f ig ab rt d o f f - - level 24
The service will be stopped the next time you enter one of these runlevels. If you need to stop the
service immediately, use the service command as described in Section 11.3.3, “ Stopping a Service” .
D o not use the - - level option when working with a service that is managed by xin et d , as it is not
supported. For example, to disable the rsyn c service, type:
~]# ch kco n f ig rsyn c o f f
If the xin et d daemon is running, the service is immediately disabled without having to manually
restart the daemon.

11.3. Running Services
The service utility allows you to start, stop, or restart the services from the /et c/in it .d / directory.

11.3.1. Det ermining t he Service St at us
To determine the current status of a service, type the following at a shell prompt:
service service_name st at u s
For example, to determine the status of the h t t p d service, type:

199

Deployment G uide

~]# service h t t p d st at u s
httpd (pid 7474) is running...
To display the status of all available services at once, run the service command with the - - st at u sall option:
~]# service - - st at u s- all
abrt (pid 1492) is running...
acpid (pid 1305) is running...
atd (pid 1540) is running...
auditd (pid 1103) is running...
automount (pid 1315) is running...
Avahi daemon is running
cpuspeed is stopped
... several lines omitted ...
wpa_supplicant (pid 1227) is running...
Note that you can also use the Service C o n f ig u rat io n utility as described in Section 11.2.1, “ Using
the Service Configuration Utility” .

11.3.2. St art ing a Service
To start a service, type the following at a shell prompt as ro o t :
service service_name st art
For example, to start the h t t p d service, type:
~]# service h t t p d st art
Starting httpd:

[ OK ]

11.3.3. St opping a Service
To stop a running service, type the following at a shell prompt as ro o t :
service service_name st o p
For example, to stop the h t t p d service, type:
~]# service h t t p d st o p
Stopping httpd:

[ OK ]

11.3.4 . Rest art ing a Service
To restart the service, type the following at a shell prompt as ro o t :
service service_name rest art
For example, to restart the h t t p d service, type:

200

⁠Chapt er 1 1 . Services and Daemons

~]# service h t t p d rest art
Stopping httpd:
Starting httpd:

[ OK ]
[ OK ]

11.4 . Addit ional Resources
11.4 .1. Inst alled Document at ion
ch kco n f ig (8) — a manual page for the ch kco n f ig utility.
n t sysv(8) — a manual page for the n t sysv utility.
service(8) — a manual page for the service utility.
syst em- co n f ig - services(8) — a manual page for the syst em- co n f ig - services utility.

11.4 .2. Relat ed Books
Red Hat Enterprise Linux 6 Security Guide
A guide to securing Red Hat Enterprise Linux 6. It contains valuable information on how to
set up the firewall, as well as the configuration of SELin u x.

201

Deployment G uide

Chapter 12. Configuring Authentication
Authentication is the way that a user is identified and verified to a system. The authentication process
requires presenting some sort of identity and credentials, like a user name and password. The
credentials are then compared to information stored in some data store on the system. In Red Hat
Enterprise Linux, the Authentication Configuration Tool helps configure what kind of data store to
use for user credentials, such as LD AP.
For convenience and potentially part of single sign-on, Red Hat Enterprise Linux can use a central
daemon to store user credentials for a number of different data stores. The System Security Services
D aemon (SSSD ) can interact with LD AP, Kerberos, and external applications to verify user
credentials. The Authentication Configuration Tool can configure SSSD along with NIS, Winbind,
and LD AP, so that authentication processing and caching can be combined.

12.1. Configuring Syst em Aut hent icat ion
When a user logs into a Red Hat Enterprise Linux system, that user presents some sort of credential to
establish the user identity. The system then checks those credentials against the configured
authentication service. If the credentials match and the user account is active, then the user is
authenticated. (Once a user is authenticated, then the information is passed to the access control
service to determine what the user is permitted to do. Those are the resources the user is authorized to
access.)
The information to verify the user can be located on the local system or the local system can
reference a user database on a remote system, such as LD AP or Kerberos.
The system must have a configured list of valid account databases for it to check for user
authentication. On Red Hat Enterprise Linux, the Authentication Configuration Tool has both GUI
and command-line options to configure any user data stores.
A local system can use a variety of different data stores for user information, including Lightweight
D irectory Access Protocol (LD AP), Network Information Service (NIS), and Winbind. Additionally,
both LD AP and NIS data stores can use Kerberos to authenticate users.

Important
If a medium or high security level is set during installation or with the Security Level
Configuration Tool, then the firewall prevents NIS authentication. For more information about
firewalls, see the " Firewalls" section of the Security Guide.

12.1.1. Launching t he Aut hent icat ion Configurat ion T ool UI
1. Log into the system as root.
2. Open the Syst em.
3. Select the Ad min ist rat io n menu.
4. Select the Au t h en t icat io n item.

202

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Alternatively, run the syst em- co n f ig - au t h en t icat io n command.

Important
Any changes take effect immediately when the Authentication Configuration Tool UI is closed.
There are two configuration tabs in the Au t h en t icat io n dialog box:
Id en t it y & Au t h en t icat io n , which configures the resource used as the identity store (the data
repository where the user ID s and corresponding credentials are stored).
Ad van ced O p t io n s, which allows authentication methods other than passwords or certificates,
like smart cards and fingerprint.

12.1.2. Select ing t he Ident it y St ore for Aut hent icat ion
The Id en t it y & Au t h en t icat io n tab sets how users should be authenticated. The default is to use
local system authentication, meaning the users and their passwords are checked against local
system accounts. A Red Hat Enterprise Linux machine can also use external resources which contain
the users and credentials, including LD AP, NIS, and Winbind.

203

Deployment G uide

Fig u re 12.1. Lo cal Au t h en t icat io n

1 2 .1 .2 .1 . Co nfiguring LDAP Aut he nt icat io n
Either the openldap-clients package or the sssd package is used to configure an LD AP server for the
user database. Both packages are installed by default.
1. Open the Authentication Configuration Tool, as in Section 12.1.1, “ Launching the
Authentication Configuration Tool UI” .
2. Select LD AP in the U ser Acco u n t D at ab ase drop-down menu.

204

⁠Chapt er 1 2 . Configuring Aut hent icat ion

3. Set the information that is required to connect to the LD AP server.
LD AP Search B ase D N gives the root suffix or distinguished name (D N) for the user
directory. All of the user entries used for identity/authentication will exist below this parent
entry. For example, ou=people,dc=example,dc=com.
This field is optional. If it is not specified, then the System Security Services D aemon
(SSSD ) attempts to detect the search base using the namingContexts and
defaultNamingContext attributes in the LD AP server's configuration entry.
LD AP Server gives the URL of the LD AP server. This usually requires both the host name
and port number of the LD AP server, such as ldap://ldap.example.com:389.
Entering the secure protocol in the URL, ld ap s://, enables the D o wn lo ad C A C ert if icat e
button.
U se T LS t o en cryp t co n n ect io n s sets whether to use Start TLS to encrypt the
connections to the LD AP server. This enables a secure connection over a standard port.

205

Deployment G uide

Selecting TLS enables the D o wn lo ad C A C ert if icat e button, which retrieves the issuing
CA certificate for the LD AP server from whatever certificate authority issued it. The CA
certificate must be in the privacy enhanced mail (PEM) format.

Important
Do not select U se T LS t o en cryp t co n n ect io n s if the server URL uses a secure
protocol (ld ap s). This option uses Start TLS, which initiates a secure connection
over a standard port; if a secure port is specified, then a protocol like SSL must be
used instead of Start TLS.
4. Select the authentication method. LD AP allows simple password authentication or Kerberos
authentication.
Using Kerberos is described in Section 12.1.2.4, “ Using Kerberos with LD AP or NIS
Authentication” .
The LD AP p asswo rd option uses PAM applications to use LD AP authentication. This option
requires either a secure (ld ap s://) URL or the TLS option to connect to the LD AP server.

1 2 .1 .2 .2 . Co nfiguring NIS Aut he nt icat io n
1. Install the ypbind package. This is required for NIS services, but is not installed by default.
~]# yum install ypbind
When the yp b in d service is installed, the p o rt map and yp b in d services are started and
enabled to start at boot time.
2. Open the Authentication Configuration Tool, as in Section 12.1.1, “ Launching the
Authentication Configuration Tool UI” .
3. Select N IS in the U ser Acco u n t D at ab ase drop-down menu.

206

⁠Chapt er 1 2 . Configuring Aut hent icat ion

4. Set the information to connect to the NIS server, meaning the NIS domain name and the
server host name. If the NIS server is not specified, the au t h co n f ig daemon scans for the
NIS server.
5. Select the authentication method. NIS allows simple password authentication or Kerberos
authentication.
Using Kerberos is described in Section 12.1.2.4, “ Using Kerberos with LD AP or NIS
Authentication” .
For more information about NIS, see the " Securing NIS" section of the Security Guide.

1 2 .1 .2 .3. Co nfiguring Winbind Aut he nt icat io n
1. Install the samba-winbind package. This is required for Windows integration features in Samba
services, but is not installed by default.
~]# yum install samba-winbind
2. Open the Authentication Configuration Tool, as in Section 12.1.1, “ Launching the
Authentication Configuration Tool UI” .

207

Deployment G uide

3. Select Win b in d in the U ser Acco u n t D at ab ase drop-down menu.

4. Set the information that is required to connect to the Microsoft Active D irectory domain
controller.
Win b in d D o main gives the Windows domain to connect to.
This should be in the Windows 2000 format, such as D O MAIN .
Secu rit y Mo d el sets the security model to use for Samba clients. au t h co n f ig supports
four types of security models:

208

⁠Chapt er 1 2 . Configuring Aut hent icat ion

ad s configures Samba to act as a domain member in an Active D irectory Server realm.
To operate in this mode, the krb5-server package must be installed and Kerberos must
be configured properly. Also, when joining to the Active D irectory Server using the
command line, the following command must be used:
n et ad s jo in
d o main has Samba validate the user name/password by authenticating it through a
Windows primary or backup domain controller, much like a Windows server.
server has a local Samba server validate the user name/password by authenticating it
through another server, such as a Windows server. If the server authentication attempt
fails, the system then attempts to authenticate using u ser mode.
u ser requires a client to log in with a valid user name and password. This mode does
support encrypted passwords.
The user name format must be domain\user, such as EXAMPLE\jsmit h .

Note
When verifying that a given user exists in the Windows domain, always use
Windows 2000-style formats and escape the backslash (\) character. For
example:
~]# getent passwd domain\\user D OMAIN\user:*:16777216:16777216:Name
Surname:/home/D OMAIN/user:/bin/bash

This is the default option.
Win b in d AD S R ealm gives the Active D irectory realm that the Samba server will join.
This is only used with the ad s security model.
Win b in d D o main C o n t ro llers gives the domain controller to use. For more information
about domain controllers, see Section 20.1.6.3, “ D omain Controller” .
T emp lat e Sh ell sets which login shell to use for Windows user account settings.
Allo w o f f lin e lo g in allows authentication information to be stored in a local cache. The
cache is referenced when a user attempts to authenticate to system resources while the
system is offline.
For more information about the Win b in d service, see Section 20.1.2, “ Samba D aemons and Related
Services” .
For additional information about configuring Win b in d and troubleshooting tips, see the
Knowledgebase on the Red Hat Customer Portal.
Also, the Red Hat Access Labs page includes the Win b in d Map p er utility that generates a part of
the smb .co n f file to help you connect a Red Hat Enterprise Linux to an Active D irectory.

1 2 .1 .2 .4 . Using Ke rbe ro s wit h LDAP o r NIS Aut he nt icat io n

209

Deployment G uide

Both LD AP and NIS authentication stores support Kerberos authentication methods. Using Kerberos
has a couple of benefits:
It uses a security layer for communication while still allowing connections over standard ports.
It automatically uses credentials caching with SSSD , which allows offline logins.
Using Kerberos authentication requires the krb5-libs and krb5-workstation packages.
The K erb ero s p asswo rd option from the Au t h en t icat io n Met h o d drop-down menu
automatically opens the fields required to connect to the Kerberos realm.

Fig u re 12.2. K erb ero s Field s
R ealm gives the name for the realm for the Kerberos server. The realm is the network that uses
Kerberos, composed of one or more key distribution centers (KD C) and a potentially large number
of clients.
K D C s gives a comma-separated list of servers that issue Kerberos tickets.
Ad min Servers gives a list of administration servers running the kad min d process in the realm.
Optionally, use D NS to resolve server host name and to find additional KD Cs within the realm.
For more information about Kerberos, see section " Using Kerberos" of the Red Hat Enterprise Linux 6
Managing Single Sign-On and Smart Cards guide.

12.1.3. Configuring Alt ernat ive Aut hent icat ion Feat ures

210

⁠Chapt er 1 2 . Configuring Aut hent icat ion

The Authentication Configuration Tool also configures settings related to authentication behavior,
apart from the identity store. This includes entirely different authentication methods (fingerprint scans
and smart cards) or local authentication rules. These alternative authentication options are
configured in the Ad van ced O p t io n s tab.

Fig u re 12.3. Ad van ced O p t io n s

1 2 .1 .3.1 . Using Finge rprint Aut he nt icat io n
When there is appropriate hardware available, the En ab le f in g erp rin t read er su p p o rt option
allows fingerprint scans to be used to authenticate local users in addition to other credentials.

1 2 .1 .3.2 . Se t t ing Lo cal Aut he nt icat io n Param e t e rs
There are two options in the Lo cal Au t h en t icat io n O p t io n s area which define authentication
behavior on the local system:

211

Deployment G uide

En ab le lo cal access co n t ro l instructs the /et c/secu rit y/access.co n f file to check for local
user authorization rules.
Passwo rd H ash in g Alg o rit h m sets the hashing algorithm to use to encrypt locally-stored
passwords.

1 2 .1 .3.3. Enabling Sm art Card Aut he nt icat io n
When there are appropriate smart card readers available, a system can accept smart cards (or
tokens) instead of other user credentials to authenticate.
Once the En ab le smart card su p p o rt option is selected, then the behaviors of smart card
authentication can be defined:
C ard R emo val Act io n tells the system how to respond when the card is removed from the card
reader during an active session. A system can either ignore the removal and allow the user to
access resources as normal, or a system can immediately lock until the smart card is supplied.
R eq u ire smart card lo g in sets whether a smart card is required for logins or allowed for logins.
When this option is selected, all other methods of authentication are immediately blocked.

Warning
D o not select this option until you have successfully authenticated to the system using a
smart card.
Using smart cards requires the pam_pkcs11 package.

1 2 .1 .3.4 . Cre at ing Use r Ho m e Dire ct o rie s
There is an option (C reat e h o me d irect o ries o n t h e f irst lo g in ) to create a home directory
automatically the first time that a user logs in.
This option is beneficial with accounts that are managed centrally, such as with LD AP. However, this
option should not be selected if a system like automount is used to manage user home directories.

12.1.4 . Configuring Aut hent icat ion from t he Command Line
The au t h co n f ig command-line tool updates all of the configuration files and services required for
system authentication, according to the settings passed to the script. Along with allowing all of the
identity and authentication configuration options that can be set through the UI, the au t h co n f ig
tool can also be used to create backup and kickstart files.
For a complete list of au t h co n f ig options, check the help output and the man page.

1 2 .1 .4 .1 . T ips fo r Using aut hco nfig
There are some things to remember when running au t h co n f ig :
With every command, use either the - - u p d at e or - - t est option. One of those options is required
for the command to run successfully. Using - - u p d at e writes the configuration changes. - - t est
prints the changes to stdout but does not apply the changes to the configuration.
Each enable option has a corresponding disable option.

212

⁠Chapt er 1 2 . Configuring Aut hent icat ion

1 2 .1 .4 .2 . Co nfiguring LDAP Use r St o re s
To use an LD AP identity store, use the - - en ab leld ap . To use LD AP as the authentication source,
use - - en ab leld ap au t h and then the requisite connection information, like the LD AP server name,
base D N for the user suffix, and (optionally) whether to use TLS. The au t h co n f ig command also
has options to enable or disable RFC 2307bis schema for user entries, which is not possible through
the Authentication Configuration UI.
Be sure to use the full LD AP URL, including the protocol (ld ap or ld ap s) and the port number. D o
not use a secure LD AP URL (ld ap s) with the - - en ab leld ap t ls option.
authconfig --enableldap --enableldapauth -ldapserver=ldap://ldap.example.com:389,ldap://ldap2.example.com:389 -ldapbasedn=" ou=people,dc=example,dc=com" --enableldaptls -ldaploadcacert=https://ca.server.example.com/caCert.crt --update
Instead of using - - ld ap au t h for LD AP password authentication, it is possible to use Kerberos with
the LD AP user store. These options are described in Section 12.1.4.5, “ Configuring Kerberos
Authentication” .

1 2 .1 .4 .3. Co nfiguring NIS Use r St o re s
To use a NIS identity store, use the - - en ab len is. This automatically uses NIS authentication, unless
the Kerberos parameters are explicitly set, so it uses Kerberos authentication (Section 12.1.4.5,
“ Configuring Kerberos Authentication” ). The only parameters are to identify the NIS server and NIS
domain; if these are not used, then the au t h co n f ig service scans the network for NIS servers.
authconfig --enablenis --nisdomain=EXAMPLE --nisserver=nis.example.com --update

1 2 .1 .4 .4 . Co nfiguring Winbind Use r St o re s
Windows domains have several different security models, and the security model used in the domain
determines the authentication configuration for the local system.
For user and server security models, the Winbind configuration requires only the domain (or
workgroup) name and the domain controller host names.
authconfig --enablewinbind --enablewinbindauth --smbsecurity=user|server -enablewinbindoffline --smbservers=ad.example.com --smbworkgroup=EXAMPLE --update

Note
The user name format must be domain\user, such as EXAMPLE\jsmit h .
When verifying that a given user exists in the Windows domain, always use Windows 2000style formats and escape the backslash (\) character. For example:
~]# getent passwd domain\\user D OMAIN\user:*:16777216:16777216:Name
Surname:/home/D OMAIN/user:/bin/bash

213

Deployment G uide

For ads and domain security models, the Winbind configuration allows additional configuration for
the template shell and realm (ads only). For example:
authconfig --enablewinbind --enablewinbindauth --smbsecurity ads --enablewinbindoffline -smbservers=ad.example.com --smbworkgroup=EXAMPLE --smbrealm EXAMPLE.COM -winbindtemplateshell=/bin/sh --update
There are a lot of other options for configuring Windows-based authentication and the information
for Windows user accounts, such as name formats, whether to require the domain name with the user
name, and UID ranges. These options are listed in the au t h co n f ig help.

1 2 .1 .4 .5 . Co nfiguring Ke rbe ro s Aut he nt icat io n
Both LD AP and NIS allow Kerberos authentication to be used in place of their native authentication
mechanisms. At a minimum, using Kerberos authentication requires specifying the realm, the KD C,
and the administrative server. There are also options to use D NS to resolve client names and to find
additional admin servers.
authconfig NIS or LDAP options --enablekrb5 --krb5realm EXAMPLE --krb5kdc
kdc.example.com:88,server.example.com:88 --krb5adminserver server.example.com:749 -enablekrb5kdcdns --enablekrb5realmdns --update

1 2 .1 .4 .6 . Co nfiguring Lo cal Aut he nt icat io n Se t t ings
The Authentication Configuration Tool can also control some user settings that relate to security,
such as creating home directories, setting password hash algorithms, and authorization. These
settings are done independently of identity/user store settings.
For example, to create user home directories:
authconfig --enablemkhomedir --update
To set or change the hash algorithm used to encrypt user passwords:
authconfig --passalgo=sha512 --update

1 2 .1 .4 .7 . Co nfiguring Finge rprint Aut he nt icat io n
There is one option to enable support for fingerprint readers. This option can be used alone or in
conjunction with other au t h co n f ig settings, like LD AP user stores.
~]# authconfig --enablefingerprint --update

1 2 .1 .4 .8 . Co nfiguring Sm art Card Aut he nt icat io n
All that is required to use smart cards with a system is to set the - - en ab lesmart card option:
~]# authconfig --enablesmartcard --update
There are other configuration options for smart cards, such as changing the default smart card
module, setting the behavior of the system when the smart card is removed, and requiring smart
cards for login.

214

⁠Chapt er 1 2 . Configuring Aut hent icat ion

For example, this command instructs the system to lock out a user immediately if the smart card is
removed (a setting of 1 ignores it if the smart card is removed):
~]# authconfig --enablesmartcard --smartcardaction=0 --update
Once smart card authentication has been successfully configured and tested, then the system can be
configured to require smart card authentication for users rather than simple password-based
authentication.
~]# authconfig --enablerequiresmartcard --update

Warning
D o not use the - - en ab lereq u iresmart card option until you have successfully authenticated
to the system using a smart card. Otherwise, users may be unable to log into the system.

1 2 .1 .4 .9 . Managing Kickst art and Co nfigurat io n File s
The - - u p d at e option updates all of the configuration files with the configuration changes. There are
a couple of alternative options with slightly different behavior:
- - kickst art writes the updated configuration to a kickstart file.
- - t est prints the full configuration, with changes, to stdout but does not edit any configuration
files.
Additionally, au t h co n f ig can be used to back up and restore previous configurations. All archives
are saved to a unique subdirectory in the /var/lib /au t h co n f ig / directory. For example, the - saveb acku p option gives the backup directory as 2011- 07- 01:
~]# authconfig --savebackup=2011-07-01
This backs up all of the authentication configuration files beneath the
/var/lib /au t h co n f ig /b acku p - 2011- 07- 01 directory.
Any of the saved backups can be used to restore the configuration using the - - rest o reb acku p
option, giving the name of the manually-saved configuration:
~]# authconfig --restorebackup=2011-07-01
Additionally, au t h co n f ig automatically makes a backup of the configuration before it applies any
changes (with the - - u p d at e option). The configuration can be restored from the most recent
automatic backup, without having to specify the exact backup, using the - - rest o relast b acku p
option.

12.1.5. Using Cust om Home Direct ories
If LD AP users have home directories that are not in /h o me and the system is configured to create
home directories the first time users log in, then these directories are created with the wrong
permissions.

215

Deployment G uide

1. Apply the correct SELinux context and permissions from the /h o me directory to the home
directory that is created on the local system. For example:
~]# semanage fcontext -a -e /home /home/locale
2. Install the oddjob-mkhomedir package on the system.
This package provides the p am_o d d jo b _mkh o med ir.so library, which the Authentication
Configuration Tool uses to create home directories. The p am_o d d jo b _mkh o med ir.so
library, unlike the default p am_mkh o med ir.so library, can create SELinux labels.
The Authentication Configuration Tool automatically uses the
p am_o d d jo b _mkh o med ir.so library if it is available. Otherwise, it will default to using
p am_mkh o med ir.so .
3. Make sure the o d d jo b d service is running.
4. Re-run the Authentication Configuration Tool and enable home directories, as in
Section 12.1.3, “ Configuring Alternative Authentication Features” .
If home directories were created before the home directory configuration was changed, then correct
the permissions and SELinux contexts. For example:
~]# semanage fcontext -a -e /home /home/locale
# restorecon -R -v /home/locale

12.2. Using and Caching Credent ials wit h SSSD
The System Security Services D aemon (SSSD ) provides access to different identity and
authentication providers.

12.2.1. About SSSD
Most system authentication is configured locally, which means that services must check with a local
user store to determine users and credentials. What SSSD does is allow a local service to check with
a local cache in SSSD , but that cache may be taken from any variety of remote identity providers —
an LD AP directory, an Identity Management domain, Active D irectory, possibly even a Kerberos
realm.
SSSD also caches those users and credentials, so if the local system or the identity provider go
offline, the user credentials are still available to services to verify.
SSSD is an intermediary between local clients and any configured data store. This relationship
brings a number of benefits for administrators:
Reducing the load on identification/authentication servers. Rather than having every client service
attempt to contact the identification server directly, all of the local clients can contact SSSD which
can connect to the identification server or check its cache.
Permitting offline authentication. SSSD can optionally keep a cache of user identities and credentials
that it retrieves from remote services. This allows users to authenticate to resources successfully,
even if the remote identification server is offline or the local machine is offline.
Using a single user account. Remote users frequently have two (or even more) user accounts, such
as one for their local system and one for the organizational system. This is necessary to connect
to a virtual private network (VPN). Because SSSD supports caching and offline authentication,

216

⁠Chapt er 1 2 . Configuring Aut hent icat ion

remote users can connect to network resources by authenticating to their local machine and then
SSSD maintains their network credentials.

Ad d it io n al R eso u rces
While this chapter covers the basics of configuring services and domains in SSSD , this is not a
comprehensive resource. Many other configuration options are available for each functional area in
SSSD ; check out the man page for the specific functional area to get a complete list of options.
Some of the common man pages are listed in Table 12.1, “ A Sampling of SSSD Man Pages” . There is
also a complete list of SSSD man pages in the " See Also" section of the sssd ( 8) man page.
T ab le 12.1. A Samp lin g o f SSSD Man Pag es
Fu n ct io n al Area

Man Pag e

General Configuration
sudo Services
LD AP D omains
Active D irectory D omains

sssd.conf(8)
sssd-sudo
sssd-ldap
sssd-ad
sssd-ldap

Identity Management (IdM or IPA) D omains

sssd-ipa
sssd-ldap

Kerberos Authentication for D omains
OpenSSH Keys

sssd-krb5
sss_ssh_authorizedkeys
sss_ssh_knownhostsproxy

Cache Maintenance

sss_cache (cleanup)
sss_useradd, sss_usermod, sss_userdel,
sss_seed (user cache entry management)

12.2.2. Set t ing up t he sssd.conf File
SSSD services and domains are configured in a .co n f file. By default, this is /et c/sssd /sssd .co n f
— although that file must be created and configured manually, since SSSD is not configured after
installation.

1 2 .2 .2 .1 . Cre at ing t he sssd.co nf File
There are three parts of the SSSD configuration file:
[ sssd ] , for general SSSD process and operational configuration; this basically lists the
configured services, domains, and configuration parameters for each
[service_name], for configuration options for each supported system service, as described in
Section 12.2.4, “ SSSD and System Services”
[domain_type/DOMAIN_NAME], for configuration options for each configured identity provider

217

Deployment G uide

Important
While services are optional, at least one identity provider domain must be configured
before the SSSD service can be started.

Examp le 12.1. Simp le sssd .co n f File
[sssd]
domains = LOCAL
services = nss
config_file_version = 2
[nss]
filter_groups = root
filter_users = root
[domain/LOCAL]
id_provider = local
auth_provider = local
access_provider = permit

The [ sssd ] section has three important parameters:
d o main s lists all of the domains, configured in the sssd .co n f , which SSSD uses as identity
providers. If a domain is not listed in the d o main s key, it is not used by SSSD , even if it has a
configuration section.
services lists all of the system services, configured in the sssd .co n f , which use SSSD ; when
SSSD starts, the corresponding SSSD service is started for each configured system service. If a
service is not listed in the services key, it is not used by SSSD , even if it has a configuration
section.
co n f ig _f ile_versio n sets the version of the configuration file to set file format expectations. This
is version 2, for all recent SSSD versions.

Note
Even if a service or domain is configured in the sssd .co n f file, SSSD does not interact with
that service or domain unless it is listed in the services or d o main s parameters, respectively,
in the [ sssd ] section.
Other configuration parameters are listed in the sssd .co n f man page.
Each service and domain parameter is described in its respective configuration section in this
chapter and in their man pages.

1 2 .2 .2 .2 . Using a Cust o m Co nfigurat io n File
By default, the sssd process assumes that the configuration file is /et c/sssd /sssd .co n f .
An alternative file can be passed to SSSD by using the - c option with the sssd command:

218

⁠Chapt er 1 2 . Configuring Aut hent icat ion

~]# sssd -c /etc/sssd/customfile.conf --daemon

12.2.3. St art ing and St opping SSSD

Important
Configure at least one domain before starting SSSD for the first time. See Section 12.2.10,
“ SSSD and Identity Providers (D omains)” .
Either the service command or the /et c/in it .d /sssd script can start SSSD . For example:
~]# service sssd start
By default, SSSD is not configured to start automatically. There are two ways to change this
behavior:
Enabling SSSD through the au t h co n f ig command:
~]# authconfig --enablesssd --enablesssdauth --update
Adding the SSSD process to the start list using the ch kco n f ig command:
~]# chkconfig sssd on

12.2.4 . SSSD and Syst em Services
SSSD and its associated services are configured in the sssd .co n f file. The [ sssd ] section also
lists the services that are active and should be started when sssd starts within the services
directive.
SSSD can provide credentials caches for several system services:
A Name Service Switch (NSS) provider service that answers name service requests from the
sssd _n ss module. This is configured in the [ n ss] section of the SSSD configuration.
This is described in Section 12.2.5, “ Configuring Services: NSS” .
A PAM provider service that manages a PAM conversation through the sssd _p am module. This
is configured in the [ p am] section of the configuration.
This is described in Section 12.2.6, “ Configuring Services: PAM” .
An SSH provider service that defines how SSSD manages the kn o wn _h o st s file and other keyrelated configuration. Using SSSD with OpenSSH is described in Section 12.2.9, “ Configuring
Services: OpenSSH and Cached Keys” .
An au t o f s provider service that connects to an LD AP server to retrieve configured mount
locations. This is configured as part of an LD AP identity provider in a [ d o main /N AME] section in
the configuration file.
This is described in Section 12.2.7, “ Configuring Services: autofs” .

219

Deployment G uide

A su d o provider service that connects to an LD AP server to retrieve configured su d o policies.
This is configured as part of an LD AP identity provider in a [ d o main /N AME] section in the
configuration file.
This is described in Section 12.2.8, “ Configuring Services: sudo” .
A PAC responder service that defines how SSSD works with Kerberos to manage Active D irectory
users and groups. This is specifically part of managing Active D irectory identity providers with
domains, as described in Section 12.2.13, “ Creating D omains: Active D irectory” .

12.2.5. Configuring Services: NSS
SSSD provides an NSS module, sssd _n ss, which instructs the system to use SSSD to retrieve user
information. The NSS configuration must include a reference to the SSSD module, and then the
SSSD configuration sets how SSSD interacts with NSS.

Abo ut NSS Se rvice Maps and SSSD
The Name Service Switch (NSS) provides a central configuration for services to look up a number of
configuration and name resolution services. NSS provides one method of mapping system identities
and services with configuration sources.
SSSD works with NSS as a provider services for several types of NSS maps:
Passwords (p asswd )
User groups (sh ad o w)
Groups (g ro u p s)
Netgroups (n et g ro u p s)
Services (services)
Pro ced u re 12.1. C o n f ig u rin g N SS Services t o U se SSSD
NSS can use multiple identity and configuration providers for any and all of its service maps. The
default is to use system files for services; for SSSD to be included, the n ss_sss module has to be
included for the desired service type.
1. Use the Authentication Configuration tool to enable SSSD . This automatically configured the
n sswit ch .co n f file to use SSSD as a provider.
~]# authconfig --enablesssd --update
This automatically configures the password, shadow, group, and netgroups services maps to
use the SSSD module:
passwd: files sss
shadow: files sss
group:
files sss
netgroup: files sss
2. The services map is not enabled by default when SSSD is enabled with au t h co n f ig . To
include that map, open the n sswit ch .co n f file and add the sss module to the services
map:

220

⁠Chapt er 1 2 . Configuring Aut hent icat ion

~]# vim /etc/nsswitch.conf
...
services: file sss
...
Pro ced u re 12.2. C o n f ig u rin g SSSD t o Wo rk wit h N SS
The options and configuration that SSSD uses to service NSS requests are configured in the SSSD
configuration file, in the [ n ss] services section.
1. Open the sssd .co n f file.
~]# vim /etc/sssd/sssd.conf
2. Make sure that NSS is listed as one of the services that works with SSSD .
[sssd]
config_file_version = 2
reconnection_retries = 3
sbus_timeout = 30
services = n ss, pam
3. In the [ n ss] section, change any of the NSS parameters. These are listed in Table 12.2,
“ SSSD [nss] Configuration Parameters” .
[nss]
filter_groups = root
filter_users = root
reconnection_retries = 3
entry_cache_timeout = 300
entry_cache_nowait_percentage = 75
4. Restart SSSD .
~]# service sssd restart
T ab le 12.2. SSSD [ n ss] C o n f ig u rat io n Paramet ers
Paramet er

Valu e Fo rmat

D escrip t io n

221

Deployment G uide

Paramet er

Valu e Fo rmat

D escrip t io n

entry_cache_nowait_percentag
e

integer

Specifies how long sssd _n ss
should return cached entries
before refreshing the cache.
Setting this to zero (0) disables
the entry cache refresh.
This configures the entry cache
to update entries in the
background automatically if
they are requested if the time
before the next update is a
certain percentage of the next
interval. For example, if the
interval is 300 seconds and the
cache percentage is 75, then
the entry cache will begin
refreshing when a request
comes in at 225 seconds —
75% of the interval.
The allowed values for this
option are 0 to 99, which sets
the percentage based on the
en t ry_cach e_t imeo u t value.
The default value is 50% .

entry_negative_timeout

integer

filter_users, filter_groups

string

filter_users_in_groups

Boolean

debug_level

integer, 0 - 9

Specifies how long, in seconds,
sssd _n ss should cache
negative cache hits. A negative
cache hit is a query for an
invalid database entries,
including non-existent entries.
Tells SSSD to exclude certain
users from being fetched from
the NSS database. This is
particularly useful for system
accounts such as ro o t .
Sets whether users listed in the
f ilt er_u sers list appear in
group memberships when
performing group lookups. If
set to FALSE, group lookups
return all users that are
members of that group. If not
specified, this value defaults to
t ru e, which filters the group
member lists.
Sets a debug logging level.

NSS Co m pat ibilit y Mo de
NSS compatibility (compat) mode provides the support for additional entries in the /et c/p asswd file to
ensure that users or members of netgroups have access to the system.

222

⁠Chapt er 1 2 . Configuring Aut hent icat ion

To enable NSS compatibility mode to work with SSSD , add the following entries to the
/et c/n sswit ch .co n f file:
passwd: compat
passwd_compat: sss
Once NSS compatibility mode is enabled, the following p asswd entries are supported:
+ user - user
Include (+ ) or exclude (- ) a specified user from the Network Information System (NIS) map.
+ @ netgroup - @ netgroup
Include (+ ) or exclude (- ) all users in the given netgroup from the NIS map.
+
Exclude all users, except previously excluded ones from the NIS map.
For more information about NSS compatibility mode, see the n sswit ch .co n f ( 5) manual page.

12.2.6. Configuring Services: PAM

Warning
A mistake in the PAM configuration file can lock users out of the system completely. Always
back up the configuration files before performing any changes, and keep a session open so
that any changes can be reverted.
SSSD provides a PAM module, sssd _p am, which instructs the system to use SSSD to retrieve user
information. The PAM configuration must include a reference to the SSSD module, and then the
SSSD configuration sets how SSSD interacts with PAM.
Pro ced u re 12.3. C o n f ig u rin g PAM
1. Use au t h co n f ig to enable SSSD for system authentication.
# authconfig --update --enablesssd --enablesssdauth
This automatically updates the PAM configuration to reference all of the SSSD modules:
#% PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth
required
pam_env.so
auth
sufficient pam_unix.so nullok try_first_pass
auth
requisite pam_succeed_if.so uid >= 500 quiet
au t h su f f icien t p am_sss.so u se_f irst _p ass
auth
required
pam_deny.so
account
account
account

required
sufficient
sufficient

pam_unix.so
pam_localuser.so
pam_succeed_if.so uid < 500 quiet

223

Deployment G uide

acco u n t [ d ef au lt = b ad su ccess= o k u ser_u n kn o wn = ig n o re] p am_sss.so
account required
pam_permit.so
password
password
p asswo rd
password
session
session
session
sessio n
session

requisite pam_cracklib.so try_first_pass retry=3
sufficient pam_unix.so sha512 shadow nullok try_first_pass use_authtok
su f f icien t p am_sss.so u se_au t h t o k
required
pam_deny.so

optional
pam_keyinit.so revoke
required
pam_limits.so
[success=1 default=ignore] pam_succeed_if.so service in crond quiet use_uid
su f f icien t p am_sss.so
required
pam_unix.so

These modules can be set to in clu d e statements, as necessary.
2. Open the sssd .co n f file.
# vim /etc/sssd/sssd.conf
3. Make sure that PAM is listed as one of the services that works with SSSD .
[sssd]
config_file_version = 2
reconnection_retries = 3
sbus_timeout = 30
services = nss, p am
4. In the [ p am] section, change any of the PAM parameters. These are listed in Table 12.3,
“ SSSD [pam] Configuration Parameters” .
[pam]
reconnection_retries = 3
offline_credentials_expiration = 2
offline_failed_login_attempts = 3
offline_failed_login_delay = 5
5. Restart SSSD .
~]# service sssd restart
T ab le 12.3. SSSD [ p am] C o n f ig u rat io n Paramet ers
Paramet er

Valu e Fo rmat

D escrip t io n

offline_credentials_expiration

integer

Sets how long, in days, to allow
cached logins if the
authentication provider is
offline. This value is measured
from the last successful online
login. If not specified, this
defaults to zero (0), which is
unlimited.

224

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Paramet er

Valu e Fo rmat

D escrip t io n

offline_failed_login_attempts

integer

offline_failed_login_delay

integer

Sets how many failed login
attempts are allowed if the
authentication provider is
offline. If not specified, this
defaults to zero (0), which is
unlimited.
Sets how long to prevent login
attempts if a user hits the failed
login attempt limit. If set to zero
(0), the user cannot
authenticate while the provider
is offline once he hits the failed
attempt limit. Only a successful
online authentication can reenable offline authentication. If
not specified, this defaults to
five (5).

12.2.7. Configuring Services: aut ofs
Abo ut Aut o m o unt , LDAP, and SSSD
Automount maps are commonly flat files, which define a relationship between a map, a mount
directory, and a fileserver. (Automount is described in the Storage Administration Guide.)
For example, let's say that there is a fileserver called n f s.examp le.co m which hosts the directory
p u b , and automount is configured to mount directories in the /sh ares/ directory. So, the mount
location is /sh ares/p u b . All of the mounts are listed in the au t o .mast er file, which identifies the
different mount directories and the files which configure them. The au t o .sh ares file then identifies
each file server and mount directory which goes into the /sh ares/ directory. The relationships could
be viewed like this:
auto.master
_________|__________
|
|
|
|
/shares/
auto.shares
|
|
|
nfs.example.com:pub
Every mount point, then, is defined in two different files (at a minimum): the au t o .mast er and
au t o .whatever file, and those files have to be available to each local automount process.
One way for administrators to manage that for large environments is to store the automount
configuration in a central LD AP directory, and just configure each local system to point to that LD AP
directory. That means that updates only need to be made in a single location, and any new maps are
automatically recognized by local systems.
For automount-LD AP configuration, the automount files are stored as LD AP entries, which are then
translated into the requisite automount files. Each element is then translated into an LD AP attribute.
The LD AP entries look like this:

225

Deployment G uide

# container entry
dn: cn=automount,dc=example,dc=com
objectClass: nsContainer
objectClass: top
cn: automount
# master map entry
dn: automountMapName=auto.master,cn=automount,dc=example,dc=com
objectClass: automountMap
objectClass: top
automountMapName: auto.master
# shares map entry
dn: automountMapName=auto.shares,cn=automount,dc=example,dc=com
objectClass: automountMap
objectClass: top
automountMapName: auto.shares
# shares mount point
dn:
automountKey=/shares,automountMapName=auto.master,cn=automount,dc=example,dc=com
objectClass: automount
objectClass: top
automountKey: /shares
automountInformation: auto.shares
# pub mount point
dn: automountKey=pub,automountMapName=auto.shares,cn=automount,dc=example,dc=com
objectClass: automount
objectClass: top
automountKey: pub
automountInformation: filer.example.com:/pub
description: pub
The schema elements, then, match up to the structure like this (with the RFC 2307 schema):
au t o .mast er
objectclass: automountMap
filename attribute: automountMapName
_______________________|_________________________
|
|
|
|
/sh ares/
au t o .sh ares
objectclass: automount
objectclass: automountMap
mount point name attribute: automountKey
filename attribute: automountMapName
map name attribute: automountInformation
|
|
|
n f s.examp le.co m:p u b
objectclass: automount
mount point name attribute: automountKey
fileserver attribute: automountInformation

226

⁠Chapt er 1 2 . Configuring Aut hent icat ion

au t o f s uses those schema elements to derive the automount configuration. The
/et c/sysco n f ig /au t o f s file identifies the LD AP server, directory location, and schema elements used
for automount entities:
LD AP_URI=ldap://ldap.example.com
SEARCH_BASE=" cn=automount,dc=example,dc=com"
MAP_OBJECT_CLASS=" automountMap"
ENTRY_OBJECT_CLASS=" automount"
MAP_ATTRIBUTE=" automountMapName"
ENTRY_ATTRIBUTE=" automountKey"
VALUE_ATTRIBUTE=" automountInformation"
Rather than pointing the automount configuration to the LD AP directory, it can be configured to point
to SSSD . SSSD , then, stores all of the information that automount needs, and as a user attempts to
mount a directory, that information is cached into SSSD . This offers several advantages for
configuration — such as failover, service discovery, and timeouts — as well as performance
improvements by reducing the number of connections to the LD AP server. Most important, using
SSSD allows all mount information to be cached, so that clients can still successfully mount
directories even if the LDAP server goes offline.
Pro ced u re 12.4 . C o n f ig u rin g au t o f s Services in SSSD
1. Make sure that the autofs and sssd-common packages are installed.
2. Open the sssd .co n f file.
~]# vim /etc/sssd/sssd.conf
3. Add the au t o f s service to the list of services that SSSD manages.
[sssd]
services = nss,pam,au t o f s
....
4. Create a new [ au t o f s] service configuration section. This section can be left blank; there is
only one configurable option, for timeouts for negative cache hits.
This section is required, however, for SSSD to recognize the au t o f s service and supply the
default configuration.
[autofs]

5. The automount information is read from a configured LD AP domain in the SSSD
configuration, so an LD AP domain must be available. If no additional settings are made, then
the configuration defaults to the RFC 2307 schema and the LD AP search base
(ld ap _search _b ase) for the automount information. This can be customized:
The directory type, au t o f s_p ro vid er; this defaults to the id _p ro vid er value; a value of
none explicitly disables autofs for the domain.
The search base, ld ap _au t o f s_search _b ase.
The object class to use to recognize map entries, ld ap _au t o f s_map _o b ject _class

227

Deployment G uide

The attribute to use to recognize map names, ld ap _au t o f s_map _n ame
The object class to use to recognize mount point entries,
ld ap _au t o f s_en t ry_o b ject _class
The attribute to use to recognize mount point names, ld ap _au t o f s_en t ry_key
The attribute to use for additional configuration information for the mount point,
ld ap _au t o f s_en t ry_valu e
For example:
[domain/LD AP]
...
autofs_provider=ldap
ldap_autofs_search_base=cn=automount,dc=example,dc=com
ldap_autofs_map_object_class=automountMap
ldap_autofs_entry_object_class=automount
ldap_autofs_map_name=automountMapName
ldap_autofs_entry_key=automountKey
ldap_autofs_entry_value=automountInformation
6. Save and close the sssd .co n f file.
7. Configure au t o f s to look for the automount map information in SSSD by editing the
n sswit ch .co n f file and changing the location from ld ap to sss:
# vim /etc/nsswitch.conf
automount: files sss
8. Restart SSSD .
# service sssd restart

12.2.8. Configuring Services: sudo
Abo ut sudo , LDAP, and SSSD
su d o rules are defined in the su d o ers file, which must be distributed separately to every machine to
maintain consistency.
One way for administrators to manage that for large environments is to store the su d o configuration
in a central LD AP directory, and just configure each local system to point to that LD AP directory. That
means that updates only need to be made in a single location, and any new rules are automatically
recognized by local systems.
For su d o -LD AP configuration, each su d o rule is stored as an LD AP entry, with each component of
the su d o rule defined in an LD AP attribute.
The su d o ers rule looks like this:
D efaults
...
% wheel

228

env_keep+=SSH_AUTH_SOCK
ALL=(ALL)

ALL

⁠Chapt er 1 2 . Configuring Aut hent icat ion

The LD AP entry looks like this:
# sudo defaults
dn: cn=defaults,ou=SUD Oers,dc=example,dc=com
objectClass: top
objectClass: sudoRole
cn: defaults
description: D efault sudoOptions go here
sudoOption: env_keep+=SSH_AUTH_SOCK
# sudo rule
dn: cn=% wheel,ou=SUD Oers,dc=example,dc=com
objectClass: top
objectClass: sudoRole
cn: % wheel
sudoUser: % wheel
sudoHost: ALL
sudoCommand: ALL

Note
SSSD only caches su d o rules which apply to the local system, depending on the value of the
sudoHost attribute. This can mean that the sudoHost value is set to ALL, uses a regular
expression that matches the host name, matches the systems netgroup, or matches the
systems host name, fully-qualified domain name, or IP address.
The su d o service can be configured to point to an LD AP server and to pull its rule configuration
from those LD AP entries. Rather than pointing the su d o configuration to the LD AP directory, it can
be configured to point to SSSD . SSSD , then, stores all of the information that su d o needs, and
every time a user attempts a su d o -related operation, the latest su d o configuration can be pulled
from the LD AP directory (through SSSD ). SSSD , however, also caches all of the su d o riles, so that
users can perform tasks, using that centralized LD AP configuration, even if the LDAP server goes
offline.
Pro ced u re 12.5. C o n f ig u rin g su d o wit h SSSD
All of the SSSD su d o configuration options are listed in the sssd - ld ap ( 5) man page.
1. Make sure that the sssd-common package is installed.
~]$ rp m - q sssd - co mmo n
2. Open the sssd .co n f file.
~]# vim /et c/sssd /sssd .co n f
3. Add the su d o service to the list of services that SSSD manages.
[sssd]
services = nss,pam,su d o
....
4. Create a new [ su d o ] service configuration section. This section can be left blank; there is

229

Deployment G uide

only one configurable option, for evaluating the sudo not before/after period.
This section is required, however, for SSSD to recognize the su d o service and supply the
default configuration.
[sudo]
5. The su d o information is read from a configured LD AP domain in the SSSD configuration, so
an LD AP domain must be available. For an LD AP provider, these parameters are required:
The directory type, su d o _p ro vid er; this is always ld ap .
The search base, ld ap _su d o _search _b ase.
The URI for the LD AP server, ld ap _u ri.
For example:
[domain/LD AP]
id_provider = ldap
sudo_provider = ldap
ldap_uri = ldap://example.com
ldap_sudo_search_base = ou=sudoers,dc=example,dc=com
For an Identity Management (IdM or IPA) provider, there are additional parameters required to
perform Kerberos authentication when connecting to the server.
[domain/ID M]
id_provider = ipa
ipa_domain = example.com
ipa_server = ipa.example.com
ld ap _t ls_cacert = /et c/ip a/ca.crt
sudo_provider = ldap
ldap_uri = ldap://ipa.example.com
ldap_sudo_search_base = ou=sudoers,dc=example,dc=com
ld ap _sasl_mech = G SSAPI
ld ap _sasl_au t h id = h o st /h o st n ame.examp le.co m
ld ap _sasl_realm = EXAMPLE.C O M
krb 5_server = ip a.examp le.co m

Note
The su d o _p ro vid er type for an Identity Management provider is still ld ap .
6. Set the intervals to use to refresh the su d o rule cache.
The cache for a specific system user is always checked and updated whenever that user
performs a task. However, SSSD caches all rules which relate to the local system. That
complete cache is updated in two ways:
Incrementally, meaning only changes to rules since the last full update
(ld ap _su d o _smart _ref resh _in t erval, the time in seconds); the default is 15 minutes,

230

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Fully, which dumps the entire caches and pulls in all of the current rules on the LD AP
server(ld ap _su d o _f u ll_ref resh _in t erval, the time in seconds); the default is six hours.
These two refresh intervals are set separately. For example:
[domain/LD AP]
...
ldap_sudo_full_refresh_interval=86400
ldap_sudo_smart_refresh_interval=3600

Note
SSSD only caches su d o rules which apply to the local system. This can mean that the
sudoHost value is set to ALL, uses a regular expression that matches the host name,
matches the systems netgroup, or matches the systems host name, fully-qualified
domain name, or IP address.
7. Optionally, set any values to change the schema used for su d o rules.
Schema elements are set in the ld ap _su d o ru le_* attributes. By default, all of the schema
elements use the schema defined in sudoers.ldap; these defaults will be used in almost all
deployments.
8. Save and close the sssd .co n f file.
9. Configure su d o to look for rules configuration in SSSD by editing the n sswit ch .co n f file
and adding the sss location:
~]# vim /et c/n sswit ch .co n f
sudoers: files sss
10. Restart SSSD .
~]# service sssd rest art

12.2.9. Configuring Services: OpenSSH and Cached Keys
OpenSSH creates secure, encrypted connections between two systems. One machine authenticates
to another machine to allow access; the authentication can be of the machine itself for server
connections or of a user on that machine. OpenSSH is described in more detail in Chapter 13,
OpenSSH.
This authentication is performed through public-private key pairs that identify the authenticating user
or machine. The remote machine or user attempting to access the machine presents a key pair. The
local machine then elects whether to trust that remote entity; if it is trusted, the public key for that
remote machine is stored in the kn o wn _h o st s file or for the remote user in au t h o riz ed _keys.
Whenever that remote machine or user attempts to authenticate again, the local system checks the
kn o wn _h o st s or au t h o riz ed _keys file first to see if that remote entity is recognized and trusted. If it
is, then access is granted.
The first problem comes in verifying those identities reliably.
The kn o wn _h o st s file is a triplet of the machine name, its IP address, and its public key:

231

Deployment G uide

server.example.com,255.255.255.255 ssh-rsa
AbcdEfg1234Z YX098776/AbcdEfg1234Z YX098776/AbcdEfg1234Z YX098776=
The kn o wn _h o st s file can quickly become outdated for a number of different reasons: systems
using D HCP cycle through IP addresses, new keys can be re-issued periodically, or virtual machines
or services can be brought online and removed. This changes the host name, IP address, and key
triplet.
Administrators have to clean and maintain a current kn o wn _h o st s file to maintain security. (Or
system users get in the habit of accepting any machine and key presented, which negates the
security benefits of key-based security.)
Additionally, a problem for both machines and users is distributing keys in a scalable way. Machines
can send their keys as part of establishing an encrypted session, but users have to supply their keys
in advance. Simply propagating and then updating keys consistently is a difficult administrative
task.
Lastly, SSH key and machine information are only maintained locally. There may be machines or
users on the network which are recognized and trusted by some systems and not by others because
the kn o wn _h o st s file has not been updated uniformly.
The goal of SSSD is to server as a credentials cache. This includes working as a credentials cache
for SSH public keys for machines and users. OpenSSH is configured to reference SSSD to check for
cached keys; SSSD uses Red Hat Linux's Identity Management (IPA) domain as an identity, and
Identity Management actually stores the public keys and host information.

Note
Only Linux machines enrolled, or joined, in the Identity Management domain can use SSSD
as a key cache for OpenSSH. Other Unix machines and Windows machines must use the
regular authentication mechanisms with the kn o wn _h o st s file.

Co nfiguring Ope nSSH t o Use SSSD fo r Ho st Ke ys
OpenSSH is configured in either a user-specific configuration file (~ /.ssh /co n f ig ) or a system-wide
configuration file (/et c/ssh /ssh _co n f ig ). The user file has precedence over the system settings and
the first obtained value for a parameter is used. The formatting and conventions for this file are
covered in Chapter 13, OpenSSH.
In order to manage host keys, SSSD has a tool, sss_ssh _kn o wn h o st sp ro xy, which performs two
operations:
1. Asks SSSD to retrieve the public host key from the Identity Management server and store it in
the /var/lib /sss/p u b co n f /kn o wn _h o st s file.
2. Establishes a connection with the host machine, using either a socket (the default) or a proxy
command.
This tool has the format:
sss_ssh_knownhostsproxy [-d sssd_domain] [-p ssh_port] HOST [PROXY_COMMAND]
T ab le 12.4 . sss_ssh _kn o wn h o st sp ro xy O p t io n s

232

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Sh o rt Arg u men t

Lo n g Arg u men t

D escrip t io n

HOSTNAME

Gives the host name of the host
to check and connect to. In the
OpenSSH configuration file,
this can be a token, % h .
Passes a proxy command to
use to connect to the SSH
client. This is similar to running
ssh - o
Pro xyC o mman d = value. This
option is used when running
sss_ssh _kn o wn h o st sp ro xy
from the command line or
through another script, but is
not necessary in the OpenSSH
configuration file.
Only searches for public keys
in entries in the specified
domain. If not given, SSSD
searches for keys in all
configured domains.
Uses this port to connect to the
SSH client. By default, this is
port 22.

PROXY_COMMAND

-d sssd_domain

--domain sssd_domain

-p port

--port port

To use this SSSD tool, add or edit two parameters to the ssh _co n f ig or ~ /.ssh /co n f ig file:
Specify the command to use to connect to the SSH client (Pro xyC o mman d ). This is the
sss_ssh _kn o wn h o st sp ro xy, with the desired arguments and host name.
Specify the location of the SSSD hosts file (G lo b alK n o wn H o st sFile).
For example, this looks for public keys in all configured SSSD domains and connects over whatever
port and host are supplied:
ProxyCommand /usr/bin/sss_ssh_knownhostsproxy -p % p % h
GlobalKnownHostsFile /var/lib/sss/pubconf/known_hosts

Co nfiguring Ope nSSH t o Use SSSD fo r Use r Ke ys
SSSD can provide user public keys to OpenSSH. The keys are read by the SSH daemon, ssh d ,
directly from the output of the sss_ssh _au t h o riz ed keys tool and are not stored in a file.
To configure ssh d to read a user's public keys from an external program, in this case the
sss_ssh _au t h o riz ed keys tool, use the AuthorizedKeysCommand directive in the
/et c/ssh /ssh d _co n f ig file.
The sss_ssh _au t h o riz ed keys tool can be used to acquire SSH public keys from the user entries in
the Identity Management (IPA) domain and output them in OpenSSH au t h o riz ed _keys format. The
command has the following format:
sss_ssh_authorizedkeys [-d sssd_domain] USER
T ab le 12.5. sss_ssh _au t h o riz ed keys O p t io n s

233

Deployment G uide

Sh o rt Arg u men t

-d sssd_domain

Lo n g Arg u men t

D escrip t io n

USER

The user name or account
name for which to obtain the
public key. In the OpenSSH
configuration file, this can be
represented by a token, % u .
Only search for public keys in
entries in the specified domain.
If not given, SSSD searches for
keys in all configured domains.

--domain sssd_domain

This feature is configured in /et c/ssh /ssh d _co n f ig as follows:
AuthorizedKeysCommand /usr/bin/sss_ssh_authorizedkeys
AuthorizedKeysCommandRunAs nobody
These and further options are documented in the ssh d _co n f ig ( 5) man page. Note that the ssh d
service must be restarted for any changes to take effect.

12.2.10. SSSD and Ident it y Providers (Domains)
SSSD recognizes domains, which are entries within the SSSD configuration file associated with
different, external data sources. D omains are a combination of an identity provider (for user
information) and, optionally, other providers such as authentication (for authentication requests)
and for other operations, such as password changes. (The identity provider can also be used for all
operations, if all operations are performed within a single domain or server.)
SSSD works with different LD AP identity providers (including OpenLD AP, Red Hat D irectory Server,
and Microsoft Active D irectory) and can use native LD AP authentication, Kerberos authentication, or
provider-specific authentication protocols (such as Active D irectory).
A domain configuration defines the identity provider, the authentication provider, and any specific
configuration to access the information in those providers. There are several types of identity and
authentication providers:
LD AP, for general LD AP servers
Active D irectory (an extension of the LD AP provider type)
Identity Management (an extension of the LD AP provider type)
Local, for the local SSSD database
Proxy
Kerberos (authentication provider only)
The identity and authentication providers can be configured in different combinations in the domain
entry. The possible combinations are listed in Table 12.6, “ Identity Store and Authentication Type
Combinations” .
T ab le 12.6 . Id en t it y St o re an d Au t h en t icat io n T yp e C o mb in at io n s
Id en t if icat io n Pro vid er

Au t h en t icat io n Pro vid er

Identity Management (LD AP)
Active D irectory (LD AP)

Identity Management (LD AP)
Active D irectory (LD AP)

234

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Id en t if icat io n Pro vid er

Au t h en t icat io n Pro vid er

Active D irectory (LD AP)
LD AP
LD AP
proxy
proxy
proxy

Kerberos
LD AP
Kerberos
LD AP
Kerberos
proxy

Along with the domain entry itself, the domain name must be added to the list of domains that SSSD
will query. For example:
[sssd]
domains = LOCAL,Name
...
[domain/Name]
id_provider = type
auth_provider = type
provider_specific = value
global = value
global attributes are available to any type of domain, such as cache and time out settings. Each
identity and authentication provider has its own set of required and optional configuration
parameters.
T ab le 12.7. G en eral [ d o main ] C o n f ig u rat io n Paramet ers
Paramet er

Valu e Fo rmat

D escrip t io n

id_provider

string

Specifies the data back end to
use for this domain. The
supported identity back ends
are:
ldap
ipa (Identity Management in
Red Hat Enterprise Linux)
ad (Microsoft Active
D irectory)
proxy, for a legacy NSS
provider, such as n ss_n is.
Using a proxy ID provider
also requires specifying the
legacy NSS library to load
to start successfully, set in
the p ro xy_lib _n ame
option.
local, the SSSD internal
local provider

235

Deployment G uide

Paramet er

Valu e Fo rmat

D escrip t io n

auth_provider

string

min_id,max_id

integer

Sets the authentication
provider used for the domain.
The default value for this option
is the value of id _p ro vid er.
The supported authentication
providers are ldap, ipa, ad,
krb5 (Kerberos), proxy, and
none.
Optional. Specifies the UID and
GID range for the domain. If a
domain contains entries that
are outside that range, they are
ignored. The default value for
min _id is 1; the default value
for max_id is 0, which is
unlimited.

Important
The default min _id
value is the same for all
types of identity provider.
If LD AP directories are
using UID numbers that
start at one, it could
cause conflicts with
users in the local
/et c/p asswd file. To
avoid these conflicts, set
min _id to 1000 or
higher as possible.
cache_credentials

Boolean

entry_cache_timeout

integer

236

Optional. Specifies whether to
store user credentials in the
local SSSD domain database
cache. The default value for
this parameter is f alse. Set this
value to t ru e for domains other
than the LOCAL domain to
enable offline authentication.
Optional. Specifies how long, in
seconds, SSSD should cache
positive cache hits. A positive
cache hit is a successful query.

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Paramet er

Valu e Fo rmat

D escrip t io n

use_fully_qualified_names

Boolean

Optional. Specifies whether
requests to this domain require
fully-qualified domain names. If
set to t ru e, all requests to this
domain must use fully-qualified
domain names. It also means
that the output from the request
displays the fully-qualified
name. Restricting requests to
fully-qualified user names
allows SSSD to differentiate
between domains with users
with conflicting user names.
If use_fully_qualified_names is
set to f alse, it is possible to
use the fully-qualified name in
the requests, but only the
simplified version is displayed
in the output.
SSSD can only parse names
based on the domain name, not
the realm name. The same
name can be used for both
domains and realms, however.

12.2.11. Creat ing Domains: LDAP
An LD AP domain means that SSSD uses an LD AP directory as the identity provider (and, optionally,
also as an authentication provider). SSSD supports several major directory services:
Red Hat D irectory Server
OpenLD AP
Identity Management (IdM or IPA)
Microsoft Active D irectory 2008 R2

Note
All of the parameters available to a general LD AP identity provider are also available to
Identity Management and Active D irectory identity providers, which are subsets of the LD AP
provider.

Param e t e rs fo r Co nfiguring an LDAP Do m ain

237

Deployment G uide

An LD AP directory can function as both an identity provider and an authentication provider. The
configuration requires enough information to identify and connect to the user directory in the LD AP
server, but the way that those connection parameters are defined is flexible.
Other options are available to provide more fine-grained control, like specifying a user account to
use to connect to the LD AP server or using different LD AP servers for password operations. The most
common options are listed in Table 12.8, “ LD AP D omain Configuration Parameters” .

Note
Server-side password policies always take precedence over the policy enabled from the client
side. For example, when setting the ld ap _p wd _p o licy= sh ad o w option, the policies defined
with the shadow LPAD attributes for a user have no effect on whether the password policy is
enabled on the OpenLD AP server.

T ip
Many other options are listed in the man page for LD AP domain configuration, sssd - ld ap ( 5) .

T ab le 12.8. LD AP D o main C o n f ig u rat io n Paramet ers
Paramet er

D escrip t io n

ldap_uri

Gives a comma-separated list of the URIs of the
LD AP servers to which SSSD will connect. The
list is given in order of preference, so the first
server in the list is tried first. Listing additional
servers provides failover protection. This can be
detected from the D NS SRV records if it is not
given.
Gives the base D N to use for performing LD AP
user operations.

ldap_search_base

Important
If used incorrectly, ldap_search_base
might cause SSSD lookups to fail.
With an AD provider, setting ldap_search_base
is not required. The AD provider automatically
discovers all the necessary information. Red Hat
recommends not to set the parameter in this
situation and instead rely on what the AD
provider discovers.

238

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Paramet er

D escrip t io n

ldap_tls_reqcert

Specifies how to check for SSL server
certificates in a TLS session. There are four
options:
never disables requests for certificates.
allow requests a certificate, but proceeds
normally even if no certificate is given or a
bad certificate is given.
try requests a certificate and proceeds
normally if no certificate is given, If a bad
certificate is given, the session terminates.
demand and hard are the same option. This
requires a valid certificate or the session is
terminated.
The default is hard.

ldap_tls_cacert

ldap_referrals

Gives the full path and file name to the file that
contains the CA certificates for all of the CAs that
SSSD recognizes. SSSD will accept any
certificate issued by these CAs. This uses the
OpenLD AP system defaults if it is not given
explicitly.
Sets whether SSSD will use LD AP referrals,
meaning forwarding queries from one LD AP
database to another. SSSD supports databaselevel and subtree referrals. For referrals within
the same LD AP server, SSSD will adjust the D N
of the entry being queried. For referrals that go
to different LD AP servers, SSSD does an exact
match on the D N. Setting this value to t ru e
enables referrals; this is the default.
Referrals can negatively impact overall
performance because of the time spent
attempting to trace referrals. D isabling referral
checking can significantly improve performance.

239

Deployment G uide

Paramet er

D escrip t io n

ldap_schema

Sets what version of schema to use when
searching for user entries. This can be rf c2307,
rf c2307b is, ad , or ip a. The default is rf c2307.
In RFC 2307, group objects use a multi-valued
attribute, memberuid, which lists the names of
the users that belong to that group. In RFC
2307bis, group objects use the member
attribute, which contains the full distinguished
name (D N) of a user or group entry. RFC
2307bis allows nested groups using the
member attribute. Because these different
schema use different definitions for group
membership, using the wrong LD AP schema
with SSSD can affect both viewing and
managing network resources, even if the
appropriate permissions are in place.
For example, with RFC 2307bis, all groups are
returned when using nested groups or
primary/secondary groups.
$ id
uid=500(myserver) gid=500(myserver)
groups=500(myserver),510(myothergroup)
If SSSD is using RFC 2307 schema, only the
primary group is returned.
This setting only affects how SSSD determines
the group members. It does not change the
actual user data.

ldap_search_timeout

ldap_network_timeout

ldap_opt_timeout

Sets the time, in seconds, that LD AP searches
are allowed to run before they are canceled and
cached results are returned. When an LD AP
search times out, SSSD automatically switches
to offline mode.
Sets the time, in seconds, SSSD attempts to poll
an LD AP server after a connection attempt fails.
The default is six seconds.
Sets the time, in seconds, to wait before aborting
synchronous LD AP operations if no response is
received from the server. This option also
controls the timeout when communicating with
the KD C in case of a SASL bind. The default is
five seconds.

LDAP Do m ain Exam ple
The LD AP configuration is very flexible, depending on your specific environment and the SSSD
behavior. These are some common examples of an LD AP domain, but the SSSD configuration is not
limited to these examples.

24 0

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Note
Along with creating the domain entry, add the new domain to the list of domains for SSSD to
query in the sssd .co n f file. For example:
domains = LOCAL,LD AP1,AD ,PROXYNIS

Examp le 12.2. A B asic LD AP D o main C o n f ig u rat io n
An LD AP domain requires three things:
An LD AP server
The search base
A way to establish a secure connection
The last item depends on the LD AP environment. SSSD requires a secure connection since it
handles sensitive information. This connection can be a dedicated TLS/SSL connection or it can
use Start TLS.
Using a dedicated TLS/SSL connection uses an LD APS connection to connect to the server and is
therefore set as part of the ld ap _u ri option:
# An LD AP domain
[domain/LD AP]
cache_credentials = true
id_provider = ldap
auth_provider = ldap
ldap_uri = ldaps://ldap.example.com:636
ldap_search_base = dc=example,dc=com
Using Start TLS requires a way to input the certificate information to establish a secure connection
dynamically over an insecure port. This is done using the ld ap _id _u se_st art _t ls option to use
Start TLS and then ld ap _t ls_cacert to identify the CA certificate which issued the SSL server
certificates.
# An LD AP domain
[domain/LD AP]
cache_credentials = true
id_provider = ldap
auth_provider = ldap
ldap_uri = ldap://ldap.example.com
ldap_search_base = dc=example,dc=com
ldap_id_use_start_tls = true
ldap_tls_reqcert = demand
ldap_tls_cacert = /etc/pki/tls/certs/ca-bundle.crt

24 1

Deployment G uide

12.2.12. Creat ing Domains: Ident it y Management (IdM)
The Identity Management (IdM or IPA) identity provider is an extension of a generic LD AP provider.
All of the configuration options for an LD AP provider are available to the IdM provider, as well as
some additional parameters which allow SSSD to work as a client of the IdM domain and extend IdM
functionality.
Identity Management can work as a provider for identities, authentication, access control rules, and
passwords, all of the *_provider parameters for a domain. Additionally, Identity Management has
configuration options within its own domain to manage SELinux policies, automount information,
and host-based access control. All of those features in IdM domains can be tied to SSSD
configuraiton, allowing those security-related policies to be applied and cached for system users.

Examp le 12.3. B asic Id M Pro vid er
An IdM provider, like an LD AP provider, can be set to serve several different services, including
identity, authentication, and access control
For IdM servers, there are two additional settings which are very useful (although not required):
Use the specific IdM schema rather than the default RFC 2307 schema.
Set SSSD to update the Identity Management domain's D NS server with the IP address of this
client when the client first connects to the IdM domain.
[sssd]
domains = local,example.com
...
[domain/example.com]
id_provider = ipa
ipa_server = ipaserver.example.com
ipa_hostname = ipa1.example.com
auth_provider = ipa
access_provider = ipa
chpass_provider = ipa
# set which schema to use
ldap_schema = ipa
# automatically update IdM D NS records
ipa_dyndns_update = true

Identity Management defines and maintains security policies and identities for users across a Linux
domain. This includes access control policies, SELinux policies, and other rules. Some of these
elements in the IdM domain interact directly with SSSD , using SSSD as an IdM client — and those
features can be managed in the IdM domain entry in sssd .co n f .
Most of the configuration parameters relate to setting schema elements (which is not relevant in most
deployments because IdM uses a fixed schema) and never need to be changed. In fact, none of the
features in IdM require client-side settings. But there may be circumstances where tweaking the
behavior is helpful.

Examp le 12.4 . Id M Pro vid er wit h SELin u x

24 2

⁠Chapt er 1 2 . Configuring Aut hent icat ion

IdM can define SELinux user policies for system users, so it can work as an SELinux provider for
SSSD . This is set in the selin u x_p ro vid er parameter. The provider defaults to the id _p ro vid er
value, so this is not necessary to set explicitly to support SELinux rules. However, it can be useful
to explicitly disable SELinux support for the IdM provider in SSSD .
selinux_provider = ipa

Examp le 12.5. Id M Pro vid er wit h H o st - B ased Access C o n t ro l
IdM can define host-based access controls, restricting access to services or entire systems based
on what host a user is using to connect or attempting to connect to. This rules can be evaluated
and enforced by SSSD as part of the access provider behavior.
For host-based access controls to be in effect, the Identity Management server must be the access
provider, at a minimum.
There are two options which can be set for how SSSD evaluates host-based access control rules:
SSSD can evaluate what machine (source host) the user is using to connect to the IdM
resource; this is disabled by default, so that only the target host part of the rule is evaluated.
SSSD can refresh the host-based access control rules in its cache at a specified interval.
For example:
access_provider = ipa
ipa_hbac_refresh = 120
# check for source machine rules; disabled by default
ipa_hbac_support_srchost = true

Examp le 12.6 . Id en t it y Man ag emen t wit h C ro ss- R ealm K erb ero s T ru st s
Identity Management (IdM or IPA) can be configured with trusted relationships between Active
D irectory D NS domains and Kerberos realms. This allows Active D irectory users to access
services and hosts on Linux systems.
There are two configuration settings in SSSD that are used with cross-realm trusts:
A service that adds required data to Kerberos tickets
A setting to support subdomains
K erb ero s T icket D at a
Microsoft uses a special authorization structure called privileged access certificates or MS-PAC. A
PAC is embedded in a Kerberos ticket as a way of identifying the entity to other Windows clients
and servers in the Windows domain.
SSSD has a special PAC service which generates the additional data for Kerberos tickets. When
using an Active D irectory domain, it may be necessary to include the PAC data for Windows users.
In that case, enable the p ac service in SSSD :

24 3

Deployment G uide

[sssd]
services = nss, pam, p ac
...
Win d o ws Su b d o main s
Normally, a domain entry in SSSD corresponds directly to a single identity provider. However, with
IdM cross-realm trusts, the IdM domain can trust another domain, so that the domains are
transparent to each other. SSSD can follow that trusted relationship, so that if an IdM domain is
configured, any Windows domain is also automatically searched and supported by SSSD —
without having to be configured in a domain section in SSSD .
This is configured by adding the su b d o main s_p ro vid er parameter to the IdM domain section.
This is actually an optional parameter; if a subdomain is discovered, then SSSD defaults to using
the ip a provider type. However, this parameter can also be used to disable subdomain fetches by
setting a value of n o n e.
[domain/ID M]
...
subdomains_provider = ipa
get_domains_timeout = 300

12.2.13. Creat ing Domains: Act ive Direct ory
The Active D irectory identity provider is an extension of a generic LD AP provider. All of the
configuration options for an LD AP provider are available to the Active D irectory provider, as well as
some additional parameters related to user accounts and identity mapping between Active D irectory
and system users.
There are some fundamental differences between standard LD AP servers and an Active D irectory
server. When configuring an Active D irectory provider, there are some configuration areas, then,
which require specific configuration:
Identities using a Windows security ID must be mapped to the corresponding Linux system user
ID .
Searches must account for the range retrieval extension.
There may be performance issues with LD AP referrals.

Mapping Act ive Dire ct o ry Se curiy IDs and Linux Use r IDs
There are inherent structural differences between how Windows and Linux handle system users and
in the user schemas used in Active D irectory and standard LD APv3 directory services. When using
an Active D irectory identity provider with SSSD to manage system users, it is necessary to reconcile
the Active D irectory-style user to the new SSSD user. There are two ways to do this:
Using Services for Unix to insert POSIX attributes on Windows user and group entries, and then
having those attributes pulled into PAM/NSS
Using ID mapping on SSSD to create a map between Active D irectory security ID s (SID s) and the
generated UID s on Linux
ID mapping is the simplest option for most environments because it requires no additional packages
or configuration on Active D irectory.

24 4

⁠Chapt er 1 2 . Configuring Aut hent icat ion

T he Me chanism o f ID Mapping
Linux/Unix systems use a local user ID number and group ID number to identify users on the system.
These UID :GID numbers are a simple integer, such as 501:501. These numbers are simple because
they are always created and administered locally, even for systems which are part of a larger
Linux/Unix domain.
Microsoft Windows and Active D irectory use a different user ID structure to identify users, groups,
and machines. Each ID is constructed of different segments that identify the security version, the
issuing authority type, the machine, and the identity itself. For example:
S-1-5-21-3623811015-3361044348-30300820-1013
The third through sixth blocks are the machine identifier:
S-1-5-21- 36 23811015- 336 104 4 34 8- 30300820-1013
The last block is the relative identifier (RID ) which identifies the specific entity:
S-1-5-21-3623811015-3361044348-30300820-1013
A range of possible ID numbers are always assigned to SSSD . (This is a local range, so it is the
same for every machine.)
|_____________________________|
|
|
minimum ID
max ID
This range is divided into 10,000 sections (by default), with each section allocated 200,000 ID s.
| slice 1 | slice 2 | ... |
|_________|_________|_________|
|
|
|
|
minimum ID
max ID
When a new Active D irectory domain is detected, the SID is hashed. Then, SSSD takes the modulus
of the hash and the number of available sections to determine which ID section to assign to the
Active D irectory domain. This is a reliably consistent means of assigning ID sections, so the same ID
range is assigned to the same Active D irectory domain on most client machines.
| Active | Active |
|
|D irectory|D irectory|
|
|domain 1 |domain 2 | ... |
|
|
|
|
| slice 1 | slice 2 | ... |
|_________|_________|_________|
|
|
|
|
minimum ID
max ID

24 5

Deployment G uide

Note
While the method of assigning ID sections is consistent, ID map p in g is b ased o n t h e
o rd er t h at an Act ive D irect o ry d o main is en co u n t ered o n a clien t mach in e — so it
may not result in consistent ID range assignments on all Linux client machines. If consistency
is required, then consider disabling ID mapping and using explicit POSIX attributes.

ID Mapping Param e t e rs
ID mapping is enabled in two parameters, one to enable the mapping and one to load the
appropriate Active D irectory user schema:
ldap_id_mapping = True
ldap_schema = ad

Note
When ID mapping is enabled, the uidNumber and gidNumber attributes are ignored. This
prevents any manually-assigned values. If any values must be manually assigned, then all
values must be manually assigned, and ID mapping should be disabled.

Mapping Use rs
When an Active D irectory user attempts to log into a local system service for the first time, an entry for
that user is created in the SSSD cache. The remote user is set up much like a system user:
A system UID is created for the user based on his SID and the ID range for that domain.
A GID is created for the user, which is identical to the UID .
A private group is created for the user.
A home directory is created, based on the home directory format in the sssd .co n f file.
A shell is created, according to the system defaults or the setting in the sssd .co n f file.
If the user belongs to any groups in the Active D irectory domain, then, using the SID , SSSD adds
the user to those groups on the Linux system.

Act ive Dire ct o ry Use rs and Range Re t rie val Se arche s
Microsoft Active D irectory has an attribute, MaxValRange, which sets a limit on how many values for a
multi-valued attribute will be returned. This is the range retrieval search extension. Essentially, this
runs multiuple mini-searches, each returning a subset of the results within a given range, until all
matches are returned.
For example, when doing a search for the member attribute, each entry could have multiple values,
and there can be multiple entries with that attribute. If there are 2000 matching results (or more), then
MaxValRange limits how many are displayed at once; this is the value range. The given attribute then
has an additional flag set, showing which range in the set the result is in:
attribute:range=low-high:value

24 6

⁠Chapt er 1 2 . Configuring Aut hent icat ion

For example, results 100 to 500 in a search:
member;range=99-499: cn=John Smith...
This is described in the Microsoft documentation at http://msdn.microsoft.com/enus/library/cc223242.aspx.
SSSD supports range retrievals with Active D irectory providers as part of user and group
management, without any additional configuration.
However, some LD AP provider attributes which are available to configure searches — such as
ld ap _u ser_search _b ase — are not performant with range retrievals. Be cautious when configuring
search bases in the Active D irectory provider domain and consider what searches may trigger a
range retrieval.

Pe rfo rm ance and LDAP Re fe rrals
Referrals can negatively impact overall performance because of the time spent attempting to trace
referrals. There is particularly bad performance degradation when referral chasing is used with an
Active D irectory identity provider. D isabling referral checking can significantly improve performance.
LD AP referrals are enabled by default, so they must be explicitly disabled in the LD AP domain
configuration. For example:
ldap_referrals = false

Act ive Dire ct o ry as Ot he r Pro vide r T ype s
Active D irectory can be used as an identity provider and as an access, password, and
authentication provider.
There are a number of options in the generic LD AP provider configuration which can be used to
configure an Active D irectory provider. Using the ad value is a short-cut which automatically pulls in
the parameters and values to configure a given provider for Active D irectory. For example, using
access_p ro vid er = ad to configure an Active D irectory access provider expands to this
configuration using the explicit LD AP provider parameters:
access_provider = ldap
ldap_access_order = expire
ldap_account_expire_policy = ad
Pro ced u re 12.6 . C o n f ig u rin g an Act ive D irect o ry Id en t it y Pro vid er
Active D irectory can work as a provider for identities, authentication, access control rules, and
passwords, all of the *_provider parameters for a domain. Additionally, it is possible to load the
native Active D irectory schema for user and group entries, rather than using the default RFC 2307.
1. Make sure that both the Active D irectory and Linux systems have a properly configured
environment.
Name resolution must be properly configured, particularly if service discovery is used with
SSSD .
The clocks on both systems must be in sync for Kerberos to work properly.

24 7

Deployment G uide

2. Set up the Linux system as an Active D irectory client and enroll it within the Active D irectory
domain. This is done by configuring the Kerberos and Samba services on the Linux system.
a. Set up Kerberos to use the Active D irectory Kerberos realm.
i. Open the Kerberos client configuration file.
~]# vim /etc/krb5.conf
ii. Configure the [ lo g g in g ] and [ lib d ef au lt s] sections so that they connect to
the Active D irectory realm.
[logging]
default = FILE:/var/log/krb5libs.log
[libdefaults]
default_realm = EXAMPLE.COM
dns_lookup_realm = true
dns_lookup_kdc = true
ticket_lifetime = 24h
renew_lifetime = 7d
rdns = false
forwardable = false
If autodiscovery is not used with SSSD , then also configure the [ realms] and
[ d o main _realm] sections to explicitly define the Active D irectory server.
b. Configure the Samba server to connect to the Active directory server.
i. Open the Samba configuration file.
~]# vim /etc/samba/smb.conf
ii. Set the Active D irectory domain information in the [ g lo b al] section.
[global]
workgroup = EXAMPLE
client signing = yes
client use spnego = yes
kerberos method = secrets and keytab
log file = /var/log/samba/% m.log
password server = AD .EXAMPLE.COM
realm = EXAMPLE.COM
security = ads
c. Add the Linux machine to the Active D irectory domain.
i. Obtain Kerberos credentials for a Windows administrative user.
~]# kinit Administrator
ii. Add the machine to the domain using the n et command.
~]# net ads join -k
Joined 'server' to dns domain 'example.com'

24 8

⁠Chapt er 1 2 . Configuring Aut hent icat ion

This creates a new keytab file, /et c/krb 5.keyt ab .
List the keys for the system and check that the host principal is there.
~]# klist -k
3. Use au t h co n f ig to enable SSSD for system authentication.
# authconfig --update --enablesssd --enablesssdauth
4. Set the Active D irectory domain as an identity provider in the SSSD configuration, as shown
in Example 12.7, “ An Active D irectory 2008 R2 D omain” and Example 12.8, “ An Active
D irectory 2008 R2 D omain with ID Mapping” .
5. Restart the SSH service to load the new PAM configuration.
~]# service restart sshd
6. Restart SSSD after changing the configuration file.
~]# service sssd restart

Examp le 12.7. An Act ive D irect o ry 2008 R 2 D o main
~]# vim /etc/sssd/sssd.conf
[sssd]
config_file_version = 2
domains = ad.example.com
services = nss, pam
...
[domain/ad.example.com]
id_provider = ad
ad_server = ad.example.com
ad_hostname = ad.example.com
auth_provider = ad
chpass_provider = ad
access_provider = ad
# defines user/group schema type
ldap_schema = ad
# using explicit POSIX attributes in the Windows entries
ldap_id_mapping = False
# caching credentials
cache_credentials = true
# access controls
ldap_access_order = expire
ldap_account_expire_policy = ad

24 9

Deployment G uide

ldap_force_upper_case_realm = true
# performance
ldap_referrals = false

There are two parameters that are critical for ID mapping: the Active D irectory schema must be loaded
(ld ap _sch ema) and ID mapping must be explicitly enabled (ld ap _id _map p in g ).

Examp le 12.8. An Act ive D irect o ry 2008 R 2 D o main wit h ID Map p in g
~]# vim /etc/sssd/sssd.conf
[sssd]
config_file_version = 2
domains = ad.example.com
services = nss, pam
...
[domain/ad.example.com]
id_provider = ad
ad_server = ad.example.com
ad_hostname = ad.example.com
auth_provider = ad
chpass_provider = ad
access_provider = ad
# d ef in es u ser/g ro u p sch ema t yp e
ld ap _sch ema = ad
# f o r SID - U ID map p in g
ld ap _id _map p in g = T ru e
# caching credentials
cache_credentials = true
# access controls
ldap_access_order = expire
ldap_account_expire_policy = ad
ldap_force_upper_case_realm = true
# performance
ldap_referrals = false

All of the potential configuration attributes for an Active D irectory domain are listed in the sssd ld ap ( 5) and sssd - ad ( 5) man pages.

12.2.14 . Configuring Domains: Act ive Direct ory as an LDAP Provider
(Alt ernat ive)
While Active D irectory can be configured as a type-specific identity provider, it can also be
configured as a pure LD AP provider with a Kerberos authentication provider.

250

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Pro ced u re 12.7. C o n f ig u rin g Act ive D irect o ry as an LD AP Pro vid er
1. It is recommended that SSSD connect to the Active D irectory server using SASL, which means
that the local host must have a service keytab for the Windows domain on the Linux host.
This keytab can be created using Samba.
a. Configure the /et c/krb 5.co n f file to use the Active D irectory realm.
[logging]
default = FILE:/var/log/krb5libs.log
[libdefaults]
default_realm = AD .EXAMPLE.COM
dns_lookup_realm = true
dns_lookup_kdc = true
ticket_lifetime = 24h
renew_lifetime = 7d
rdns = false
forwardable = false
[realms]
# D efine only if D NS lookups are not working
# AD .EXAMPLE.COM = {
# kdc = server.ad.example.com
# admin_server = server.ad.example.com
# master_kdc = server.ad.example.com
#}
[domain_realm]
# D efine only if D NS lookups are not working
# .ad.example.com = AD .EXAMPLE.COM
# ad.example.com = AD .EXAMPLE.COM
b. Set the Samba configuration file, /et c/samb a/smb .co n f , to point to the Windows
Kerberos realm.
[global]
workgroup = EXAMPLE
client signing = yes
client use spnego = yes
kerberos method = secrets and keytab
log file = /var/log/samba/% m.log
password server = AD .EXAMPLE.COM
realm = EXAMPLE.COM
security = ads
c. To initialize Kerberos, type the following command as ro o t :
~]# kinit Administrator@EXAMPLE.COM
d. Then, run the n et ad s command to log in as an administrator principal. This
administrator account must have sufficient rights to add a machine to the Windows
domain, but it does not require domain administrator privileges.

251

Deployment G uide

~]# net ads join -U Administrator
e. Run n et ad s again to add the host machine to the domain. This can be done with the
host principal (host/FQDN) or, optionally, with the NFS service (nfs/FQDN).
~]# net ads join createupn=" host/rhel-server.example.com@AD .EXAMPLE.COM" U Administrator
2. Make sure that the Services for Unix package is installed on the Windows server.
3. Set up the Windows domain which will be used with SSSD .
a. On the Windows machine, open Server Man ag er.
b. Create the Active D irectory D omain Services role.
c. Create a new domain, such as ad .examp le.co m.
d. Add the Identity Management for UNIX service to the Active D irectory D omain Services
role. Use the Unix NIS domain as the domain name in the configuration.
4. On the Active D irectory server, create a group for the Linux users.
a. Open Ad min ist rat ive T o o ls and select Act ive D irect o ry U sers an d C o mp u t ers.
b. Select the Active D irectory domain, ad .examp le.co m.
c. In the U sers tab, right-click and select C reat e a N ew G ro u p .
d. Name the new group u n ixu sers, and save.
e. D ouble-click the u n ixu sers group entry, and open the U sers tab.
f. Open the U n ix At t rib u t es tab.
g. Set the NIS domain to the NIS domain that was configured for ad .examp le.co m and,
optionally, set a group ID (GID ) number.
5. Configure a user to be part of the Unix group.
a. Open Ad min ist rat ive T o o ls and select Act ive D irect o ry U sers an d C o mp u t ers.
b. Select the Active D irectory domain, ad .examp le.co m.
c. In the U sers tab, right-click and select C reat e a N ew U ser.
d. Name the new user ad u ser, and make sure that the U ser mu st ch an g e p asswo rd
at n ext lo g o n and Lo ck acco u n t checkboxes are not selected.
Then save the user.
e. D ouble-click the ad u ser user entry, and open the U n ix At t rib u t es tab. Make sure
that the Unix configuration matches that of the Active D irectory domain and the
u n ixg ro u p group:
The NIS domain, as created for the Active D irectory domain
The UID
The login shell, to /b in /b ash

252

⁠Chapt er 1 2 . Configuring Aut hent icat ion

The home directory, to /h o me/ad u ser
The primary group name, to u n ixu sers

T IP
Password lookups on large directories can take several seconds per request. The
initial user lookup is a call to the LD AP server. Unindexed searches are much more
resource-intensive, and therefore take longer, than indexed searches because the
server checks every entry in the directory for a match. To speed up user lookups, index
the attributes that are searched for by SSSD :
uid
uidNumber
gidNumber
gecos
6. On the Linux system, configure the SSSD domain.
~]# vim /etc/sssd/sssd.conf
For a complete list of LD AP provider parameters, see the sssd - ld ap ( 5) man pages.

Examp le 12.9 . An Act ive D irect o ry 2008 R 2 D o main wit h Services f o r U n ix
[sssd]
config_file_version = 2
domains = ad.example.com
services = nss, pam
...
[domain/ad.example.com]
cache_credentials = true
# for performance
ldap_referrals = false
id_provider = ldap
auth_provider = krb5
chpass_provider = krb5
access_provider = ldap
ldap_schema = rfc2307bis
ldap_sasl_mech = GSSAPI
ldap_sasl_authid = host/rhel-server.example.com@AD .EXAMPLE.COM
#provide the schema for services for unix
ldap_schema = rfc2307bis
ldap_user_search_base = ou=user accounts,dc=ad,dc=example,dc=com
ldap_user_object_class = user
ldap_user_home_directory = unixHomeD irectory

253

Deployment G uide

ldap_user_principal = userPrincipalName
# optional - set schema mapping
# parameters are listed in sssd-ldap
ldap_user_object_class = user
ldap_user_name = sAMAccountName
ldap_group_search_base = ou=groups,dc=ad,dc=example,dc=com
ldap_group_object_class = group
ldap_access_order = expire
ldap_account_expire_policy = ad
ldap_force_upper_case_realm = true
ldap_referrals = false
krb5_realm = AD -REALM.EXAMPLE.COM
# required
krb5_canonicalize = false

7. Restart SSSD .
~]# service sssd restart

12.2.15. Domain Opt ions: Set t ing Username Format s
One of the primary actions that SSSD performs is mapping a local system user to an identity in the
remote identity provider. SSSD uses a combination of the user name and the domain back end name
to create the login identity.
As long as they belong to different domains, SSSD can recognize different users with the same user
name. For example, SSSD can successfully authenticate both jsmit h in the ld ap .examp le.co m
domain and jsmit h in the ld ap .o t h erexamp le.co m domain.
The name format used to construct full user name is (optionally) defined universally in the [ sssd ]
section of the configuration and can then be defined individually in each domain section.
Usernames for different services — LD AP, Samba, Active D irectory, Identity Management, even the
local system — all have different formats. The expression that SSSD uses to identify user
name/domain name sets must be able to interpret names in different formats. This expression is set in
the re_exp ressio n parameter.
In the global default, this filter constructs a name in the form name@domain:
(?P[^@]+)@?(?P[^@]*$)

Note
The regular expression format is Python syntax.
The domain part may be supplied automatically, based on the domain name of the identity provider.
Therefore, a user can log in as jsmit h and if the user belongs to the LOCAL domain (for example),
then his user name is interpreted by SSSD as jsmit h @ LO C AL.

254

⁠Chapt er 1 2 . Configuring Aut hent icat ion

However, other identity providers may have other formats. Samba, for example, has a very strict
format so that user name must match the form DOMAIN\username. For Samba, then, the regular
expression must be:
(?P[^\\]*?)\\?(?P[^\\]+$)
Some providers, such as Active D irectory, support multiple different name formats. Active D irectory
and Identity Management, for example, support three different formats by default:
username
username@domain.name
DOMAIN\username
The default value for Active D irectory and Identity Management providers, then, is a more complex
filter that allows all three name formats:
(((?P[^\\]+)\\(?P.+$))|((?P[^@]+)@(?P.+$))|(^(?P
[^@\\]+)$))

Note
Requesting information with the fully-qualified name, such as jsmit h @ ld ap .examp le.co m,
always returns the proper user account. If there are multiple users with the same user name in
different domains, specifying only the user name returns the user for whichever domain comes
first in the lookup order.
While re_exp ressio n is the most important method for setting user name formats, there are two other
options which are useful for other applications.
D ef au lt D o main N ame Valu e
The first sets a default domain name to be used with all users, d ef au lt _d o main _su f f ix. (This is a
global setting, available in the [ sssd ] section only.) There may be a case where multiple domains
are configured but only one stores user data and the others are used for host or service identities.
Setting a default domain name allows users to log in with only their user name, not specifying the
domain name (which would be required for users outside the primary domain).
[sssd]
...
default_domain_suffix = USERS.EXAMPLE.COM
Fu ll N ame Fo rmat f o r O u t p u t
The other parameter is related to re_exp ressio n , only instead of defining how to interpret a user
name, it defines how to print an identified name. The f u ll_n ame_f o rmat parameter sets how the user
name and domain name (once determined) are displayed.
full_name_format = % 1$s@% 2$s

12.2.16. Domain Opt ions: Enabling Offline Aut hent icat ion

255

Deployment G uide

User identities are always cached, as well as information about the domain services. However, user
credentials are not cached by default. This means that SSSD always checks with the back end
identity provider for authentication requests. If the identity provider is offline or unavailable, there is
no way to process those authentication requests, so user authentication could fail.
It is possible to enable offline credentials caching, which stores credentials (after successful login) as
part of the user account in the SSSD cache. Therefore, even if an identity provider is unavailable,
users can still authenticate, using their stored credentials. Offline credentials caching is primarily
configured in each individual domain entry, but there are some optional settings that can be set in
the PAM service section, because credentials caching interacts with the local PAM service as well as
the remote domain.
[domain/EXAMPLE]
cache_credentials = true
There are optional parameters that set when those credentials expire. Expiration is useful because it
can prevent a user with a potentially outdated account or credentials from accessing local services
indefinitely.
The credentials expiration itself is set in the PAM service, which processes authentication requests for
the system.
[sssd]
services = nss,pam
...
[pam]
offline_credentials_expiration = 3
...
[domain/EXAMPLE]
cache_credentials = true
...
o f f lin e_cred en t ials_exp irat io n sets the number of days after a successful login that a single
credentials entry for a user is preserved in cache. Setting this to zero (0) means that entries are kept
forever.
While not related to the credentials cache specifically, each domain has configuration options on
when individual user and service caches expire:
acco u n t _cach e_exp irat io n sets the number of days after a successful login that the entire user
account entry is removed from the SSSD cache. This must be equal to or longer than the
individual offline credentials cache expiration period.
en t ry_cach e_t imeo u t sets a validity period, in seconds, for all entries stored in the cache
before SSSD requests updated information from the identity provider. There are also individual
cache timeout parameters for group, service, netgroup, sudo, and autofs entries; these are listed
in the sssd .co n f man page. The default time is 5400 seconds (90 minutes).
For example:
[sssd]
services = nss,pam
...
[pam]

256

⁠Chapt er 1 2 . Configuring Aut hent icat ion

offline_credentials_expiration = 3
...
[domain/EXAMPLE]
cache_credentials = true
account_cache_expiration = 7
entry_cache_timeout = 14400
...

12.2.17. Domain Opt ions: Set t ing Password Expirat ions
Password policies generally set an expiration time, after which passwords expire and must be
replaced. Password expiration policies are evaluated on the server side through the identity provider,
then a warning can be processed and displayed in SSSD through its PAM service.
There are two ways to display password expiration warnings:
The pam_pwd_expiration_warning parameter defines the global default setting for all domains on
how far in advance of the password expiration to display a warning. This is set for the PAM
service.
The pwd_expiration_warning parameter defines the per-domain setting on how far in advance of
the password expiration to display a warning.
When using a domain-level password expiration warning, an authentication provider
(au t h _p ro vid er) must also be configured for the domain.
For example:
[sssd]
services = nss,pam
...
[pam]
p am_p wd _exp irat io n _warn in g = 3
...
[domain/EXAMPLE]
id_provider = ipa
au t h _p ro vid er = ipa
p wd _exp irat io n _warn in g = 7
The password expiration warning must be sent from the server to SSSD for the warning to be
displayed. If no password warning is sent from the server, no message is displayed through SSSD ,
even if the password expiration time is within the period set in SSSD .
If the password expiration warning is not set in SSSD or is set to 0, then the SSSD password
warning filter is not applied and the server-side password warning is automatically displayed.

257

Deployment G uide

Note
As long as the password warning is sent from the server, the PAM or domain password
expirations in effect override the password warning settings on the back end identity provider.
For example, consider a back end identity provider that has the warning period set at 28 days,
but the PAM service in SSSD has it set to 7 days. The provider sends the warning to SSSD
starting at 28 days, but the warning is not displayed locally until 7 days, according to the
password expiration set in the SSSD configuration.

Passwo rd Expirat io n Warnings fo r No n-Passwo rd Aut he nt icat io n
By default, password expiration is verified only if the user enters the password during authentication.
However, you can configure SSSD to perform the expiration check and display the warning even
when a non-password authentication method is used, for example, during SSH login.
To enable password expiration warnings with non-password authentication methods:
1. Make sure the access_p ro vid er parameter is set to ld ap in the sssd .co n f file.
2. Make sure the ld ap _p wd _p o licy parameter is set in sssd .co n f . In most situations, the
appropriate value is sh ad o w.
3. Add one of the following p wd _exp ire_* values to the ld ap _access_o rd er parameter in
sssd .co n f . If the password is about to expire, each one of these values only displays the
expiration warning. In addition:
p wd _exp ire_p o licy_reject prevents the user from logging in if the password is already
expired.
p wd _exp ire_p o licy_warn allows the user to log in even if the password is already
expired.
p wd _exp ire_p o licy_ren ew prompts the user to immediately change the password if the
user attempts to log in with an expired password.
For example:
[domain/EXAMPLE]
access_provider = ldap
ldap_pwd_policy = shadow
ldap_access_order = pwd_expire_policy_warn
For more details on using ld ap _access_o rd er and its values, see the sssd-ldap(5) man page.

12.2.18. Domain Opt ions: Using DNS Service Discovery
D NS service discovery, defined in RFC 2782, allows applications to check the SRV records in a
given domain for certain services of a certain type; it then returns any servers discovered of that type.
With SSSD , the identity and authentication providers can either be explicitly defined (by IP address
or host name) or they can be discovered dynamically, using service discovery. If no provider server
is listed — for example, if id _p ro vid er = ld ap is set without a corresponding ld ap _u ri parameter —
then discovery is automatically used.
The D NS discovery query has this format:

258

⁠Chapt er 1 2 . Configuring Aut hent icat ion

_service._protocol.domain
For example, a scan for an LD AP server using TCP in the examp le.co m domain looks like this:
_ldap._tcp.example.com

Note
For every service with which to use service discovery, add a special D NS record to the D NS
server:
_service._protocol._domain TTL priority weight port hostname

For SSSD , the service type is LD AP by default, and almost all services use TCP (except for Kerberos,
which starts with UD P). For service discovery to be enabled, the only thing that is required is the
domain name. The default is to use the domain portion of the machine host name, but another
domain can be specified (using the d n s_d isco very_d o main parameter).
So, by default, no additional configuration needs to be made for service discovery — with one
exception. The password change provider has server discovery disabled by default, and it must be
explicitly enabled by setting a service type.
[domain/EXAMPLE]
...
chpass_provider = ldap
ldap_chpass_dns_service_name = ldap
While no configuration is necessary, it is possible for server discovery to be customized by using a
different D NS domain (d n s_d isco very_d o main ) or by setting a different service type to scan for.
For example:
[domain/EXAMPLE]
id _provider = ldap
dns_discovery_domain = corp.example.com
ldap_dns_service_name = ldap
chpass_provider = krb5
ldap_chpass_dns_service_name = kerberos
Lastly, service discovery is never used with backup servers; it is only used for the primary server for a
provider. What this means is that discovery can be used initially to locate a server, and then SSSD
can fall back to using a backup server. To use discovery for the primary server, use _srv_ as the
primary server value, and then list the backup servers. For example:
[domain/EXAMPLE]
id _provider = ldap
ldap_uri = _srv_
ldap_backup_uri = ldap://ldap2.example.com
auth_provider = krb5

259

Deployment G uide

krb5_server = _srv_
krb5_backup_server = kdc2.example.com
chpass_provider = krb5
ldap_chpass_dns_service_name = kerberos
ldap_chpass_uri = _srv_
ldap_chpass_backup_uri = kdc2.example.com

Note
Service discovery cannot be used with backup servers, only primary servers.
If a D NS lookup fails to return an IPv4 address for a host name, SSSD attempts to look up an IPv6
address before returning a failure. This only ensures that the asynchronous resolver identifies the
correct address.
The host name resolution behavior is configured in the lookup family order option in the sssd .co n f
configuration file.

12.2.19. Domain Opt ions: Using IP Addresses in Cert ificat e Subject Names
(LDAP Only)
Using an IP address in the ld ap _u ri option instead of the server name may cause the TLS/SSL
connection to fail. TLS/SSL certificates contain the server name, not the IP address. However, the
subject alternative name field in the certificate can be used to include the IP address of the server,
which allows a successful secure connection using an IP address.
Pro ced u re 12.8. U sin g IP Ad d resses in C ert if icat e Su b ject N ames
1. Convert an existing certificate into a certificate request. The signing key (- sig n key) is the key
of the issuer of whatever CA originally issued the certificate. If this is done by an external CA,
it requires a separate PEM file; if the certificate is self-signed, then this is the certificate itself.
For example:
openssl x509 -x509toreq -in old_cert.pem -out req.pem -signkey key.pem
With a self-signed certificate:
openssl x509 -x509toreq -in old_cert.pem -out req.pem -signkey old_cert.pem
2. Edit the /et c/p ki/t ls/o p en ssl.cn f configuration file to include the server's IP address under
the [ v3_ca ] section:
subjectAltName = IP:10.0.0.10
3. Use the generated certificate request to generate a new self-signed certificate with the
specified IP address:
openssl x509 -req -in req.pem -out new_cert.pem -extfile ./openssl.cnf -extensions v3_ca
-signkey old_cert.pem

260

⁠Chapt er 1 2 . Configuring Aut hent icat ion

The - ext en sio n s option sets which extensions to use with the certificate. For this, it should
be v3_ca to load the appropriate section.
4. Copy the private key block from the o ld _cert .p em file into the n ew_cert .p em file to keep all
relevant information in one file.
When creating a certificate through the cert u t il utility provided by the nss-tools package, note that
cert u t il supports D NS subject alternative names for certificate creation only.

12.2.20. Creat ing Domains: Proxy
A proxy with SSSD is just a relay, an intermediary configuration. SSSD connects to its proxy service,
and then that proxy loads the specified libraries. This allows SSSD to use some resources that it
otherwise would not be able to use. For example, SSSD only supports LD AP and Kerberos as
authentication providers, but using a proxy allows SSSD to use alternative authentication methods
like a fingerprint scanner or smart card.
T ab le 12.9 . Pro xy D o main C o n f ig u rat io n Paramet ers
Paramet er

D escrip t io n

proxy_pam_target

Specifies the target to which PAM must proxy as
an authentication provider. The PAM target is a
file containing PAM stack information in the
default PAM directory, /et c/p am.d /.
This is used to proxy an authentication
provider.

Important
Ensure that the proxy PAM stack does not
recursively include p am_sss.so .
proxy_lib_name

Specifies which existing NSS library to proxy
identity requests through. This is used to proxy
an identity provider.

Examp le 12.10. Pro xy Id en t it y an d K erb ero s Au t h en t icat io n
The proxy library is loaded using the p ro xy_lib _n ame parameter. This library can be anything as
long as it is compatible with the given authentication service. For a Kerberos authentication
provider, it must be a Kerberos-compatible library, like NIS.
[domain/PROXY_KRB5]
auth_provider = krb5
krb5_server = kdc.example.com
krb5_realm = EXAMPLE.COM
id_provider = proxy
proxy_lib_name = nis
cache_credentials = true

261

Deployment G uide

Examp le 12.11. LD AP Id en t it y an d Pro xy Au t h en t icat io n
The proxy library is loaded using the p ro xy_p am_t arg et parameter. This library must be a PAM
module that is compatible with the given identity provider. For example, this uses a PAM fingerprint
module with LD AP:
[domain/LD AP_PROXY]
id_provider = ldap
ldap_uri = ldap://example.com
ldap_search_base = dc=example,dc=com
auth_provider = proxy
proxy_pam_target = sssdpamproxy
cache_credentials = true
After the SSSD domain is configured, make sure that the specified PAM files are configured. In this
example, the target is sssd p amp ro xy, so create a /et c/p am.d /sssd p amp ro xy file and load the
PAM/LD AP modules:
auth
required
pam_frprint.so
account
required
pam_frprint.so
password
required
pam_frprint.so
session
required
pam_frprint.so

Examp le 12.12. Pro xy Id en t it y an d Au t h en t icat io n
SSSD can have a domain with both identity and authentication proxies. The only configuration
given then are the proxy settings, p ro xy_p am_t arg et for the authentication PAM module and
p ro xy_lib _n ame for the service, like NIS or LD AP.
This example illustrates a possible configuration, but this is not a realistic configuration. If LDAP is used for
identity and authentication, then both the identity and authentication providers should be set to the LDAP
configuration, not a proxy.
[domain/PROXY_PROXY]
auth_provider = proxy
id_provider = proxy
proxy_lib_name = ldap
proxy_pam_target = sssdproxyldap
cache_credentials = true
Once the SSSD domain is added, then update the system settings to configure the proxy service:
1. Create a /et c/p am.d /sssd p ro xyld ap file which requires the p am_ld ap .so module:
auth
required
pam_ldap.so
account
required
pam_ldap.so
password
required
pam_ldap.so
session
required
pam_ldap.so
2. Make sure the nss-pam-ldapd package is installed.
~]# yum install nss-pam-ldapd

262

⁠Chapt er 1 2 . Configuring Aut hent icat ion

3. Edit the /et c/n slcd .co n f file, the configuration file for the LD AP name service daemon, to
contain the information for the LD AP directory:
uid nslcd
gid ldap
uri ldaps://ldap.example.com:636
base dc=example,dc=com
ssl on
tls_cacertdir /etc/openldap/cacerts

12.2.21. Creat ing Domains: Kerberos Aut hent icat ion
Both LD AP and proxy identity providers can use a separate Kerberos domain to supply
authentication. Configuring a Kerberos authentication provider requires the key distribution center
(KD C) and the Kerberos domain. All of the principal names must be available in the specified identity
provider; if they are not, SSSD constructs the principals using the format username@REALM.

Note
Kerberos can only provide authentication; it cannot provide an identity database.
SSSD assumes that the Kerberos KD C is also a Kerberos kadmin server. However, production
environments commonly have multiple, read-only replicas of the KD C and only a single kadmin
server. Use the krb 5_kp asswd option to specify where the password changing service is running or
if it is running on a non-default port. If the krb 5_kp asswd option is not defined, SSSD tries to use
the Kerberos KD C to change the password.
The basic Kerberos configuration options are listed in Table 12.10, “ Kerberos Authentication
Configuration Parameters” . The sssd - krb 5( 5) man page has more information about Kerberos
configuration options.

Examp le 12.13. B asic K erb ero s Au t h en t icat io n
# A domain with identities provided by LD AP and authentication by Kerberos
[domain/KRBD OMAIN]
id_provider = ldap
chpass_provider = krb5
ldap_uri = ldap://ldap.example.com
ldap_search_base = dc=example,dc=com
ldap-tls_reqcert = demand
ldap_tls_cacert = /etc/pki/tls/certs/ca-bundle.crt
auth_provider = krb5
krb5_server = kdc.example.com
krb5_backup_server = kerberos.example.com
krb5_realm = EXAMPLE.COM
krb5_kpasswd = kerberos.admin.example.com
krb5_auth_timeout = 15

263

Deployment G uide

Examp le 12.14 . Set t in g K erb ero s T icket R en ewal O p t io n s
The Kerberos authentication provider, among other tasks, requests ticket granting tickets (TGT) for
users and services. These tickets are used to generate other tickets dynamically for specific
services, as accessed by the ticket principal (the user).
The TGT initially granted to the user principal is valid only for the lifetime of the ticket (by default,
whatever is configured in the configured KD C). After that, the ticket cannot be renewed or extended.
However, not renewing tickets can cause problems with some services when they try to access a
service in the middle of operations and their ticket has expired.
Kerberos tickets are not renewable by default, but ticket renewal can be enabled using the
krb 5_ren ewab le_lif et ime and krb 5_ren ew_in t erval parameters.
The lifetime for a ticket is set in SSSD with the krb 5_lif et ime parameter. This specifies how long a
single ticket is valid, and overrides any values in the KD C.
Ticket renewal itself is enabled in the krb 5_ren ewab le_lif et ime parameter, which sets the
maximum lifetime of the ticket, counting all renewals.
For example, the ticket lifetime is set at one hour and the renewable lifetime is set at 24 hours:
krb5_lifetime = 1h
krb5_renewable_lifetime = 1d
This means that the ticket expires every hour and can be renewed continually up to one day.
The lifetime and renewable lifetime values can be in seconds (s), minutes (m), hours (h), or days
(d).
The other option — which must also be set for ticket renewal — is the krb 5_ren ew_in t erval
parameter, which sets how frequently SSSD checks to see if the ticket needs to be renewed. At half
of the ticket lifetime (whatever that setting is), the ticket is renewed automatically. (This value is
always in seconds.)
krb5_lifetime = 1h
krb5_renewable_lifetime = 1d
krb5_renew_interval = 60s

Note
If the krb 5_ren ewab le_lif et ime value is not set or the krb 5_ren ew_in t erval parameter is
not set or is set to zero (0), then ticket renewal is disabled. Both krb 5_ren ewab le_lif et ime
and krb 5_ren ew_in t erval are required for ticket renewal to be enabled.

T ab le 12.10. K erb ero s Au t h en t icat io n C o n f ig u rat io n Paramet ers
Paramet er

D escrip t io n

chpass_provider

Specifies which service to use for password
change operations. This is assumed to be the
same as the authentication provider. To use
Kerberos, set this to krb5.

264

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Paramet er

D escrip t io n

krb5_server

Gives the primary Kerberos server, by IP
address or host names, to which SSSD will
connect.
Gives a comma-separated list of IP addresses or
host names of Kerberos servers to which SSSD
will connect if the primary server is not available.
The list is given in order of preference, so the
first server in the list is tried first. After an hour,
SSSD will attempt to reconnect to the primary
service specified in the krb5_server parameter.

krb5_backup_server

When using service discovery for KD C or
kpasswd servers, SSSD first searches for D NS
entries that specify UD P as the connection
protocol, and then falls back to TCP.
krb5_realm
krb5_lifetime

krb5_renewable_lifetime

krb5_renew_interval

krb5_store_password_if_offline

krb5_kpasswd

Identifies the Kerberos realm served by the KD C.
Requests a Kerberos ticket with the specified
lifetime in seconds (s), minutes (m), hours (h) or
days (d).
Requests a renewable Kerberos ticket with a
total lifetime that is specified in seconds (s),
minutes (m), hours (h) or days (d).
Sets the time, in seconds, for SSSD to check if
tickets should be renewed. Tickets are renewed
automatically once they exceed half their
lifetime. If this option is missing or set to zero,
then automatic ticket renewal is disabled.
Sets whether to store user passwords if the
Kerberos authentication provider is offline, and
then to use that cache to request tickets when
the provider is back online. The default is f alse,
which does not store passwords.
Lists alternate Kerberos kadmin servers to use if
the change password service is not running on
the KD C.

265

Deployment G uide

Paramet er

D escrip t io n

krb5_ccname_template

Gives the directory to use to store the user's
credential cache. This can be templatized, and
the following tokens are supported:
%u, the user's login name
%U, the user's login UID
%p, the user's principal name
%r, the realm name
%h, the user's home directory
%d, the value of the krb 5ccach e_d ir
parameter
%P, the process ID of the SSSD client.
%%, a literal percent sign (% )
XXXXXX, a string at the end of the template
which instructs SSSD to create a unique
filename safely
For example:
krb5_ccname_template =
FILE:% d/krb5cc_% U_XXXXXX

krb5_ccachedir

krb5_auth_timeout

Specifies the directory to store credential
caches. This can be templatized, using the same
tokens as krb 5_ccn ame_t emp lat e, except for
% d and % P. If % u , % U , % p , or % h are used,
then SSSD creates a private directory for each
user; otherwise, it creates a public directory.
Gives the time, in seconds, before an online
authentication or change password request is
aborted. If possible, the authentication request
is continued offline. The default is 15 seconds.

12.2.22. Creat ing Domains: Access Cont rol
SSSD provides a rudimentary access control for domain configuration, allowing either simple user
allow/deny lists or using the LD AP back end itself.

Using t he Sim ple Acce ss Pro vide r
The Simple Access Provider allows or denies access based on a list of user names or groups.
The Simple Access Provider is a way to restrict access to certain, specific machines. For example, if a
company uses laptops, the Simple Access Provider can be used to restrict access to only a specific
user or a specific group, even if a different user authenticated successfully against the same
authentication provider.
The most common options are simp le_allo w_u sers and simp le_allo w_g ro u p s, which grant
access explicitly to specific users (either the given users or group members) and deny access to
everyone else. It is also possible to create deny lists (which deny access only to explicit people and
implicitly allow everyone else access).
The Simple Access Provider adheres to the following four rules to determine which users should or
should not be granted access:

266

⁠Chapt er 1 2 . Configuring Aut hent icat ion

If both the allow and deny lists are empty, access is granted.
If any list is provided, allow rules are evaluated first, and then deny rules. Practically, this means
that deny rules supersede allow rules.
If an allowed list is provided, then all users are denied access unless they are in the list.
If only deny lists are provided, then all users are allowed access unless they are in the list.
This example grants access to two users and anyone who belongs to the IT group; implicitly, all
other users are denied:
[domain/example.com]
access_provider = simple
simple_allow_users = jsmith,bjensen
simple_allow_groups = itgroup

Note
The LOCAL domain in SSSD does not support simp le as an access provider.
Other options are listed in the sssd - simp le man page, but these are rarely used.

Using t he Acce ss Filt e rs
An LD AP, Active D irectory, or Identity Management server can provide access control rules for a
domain. The associated options (ld ap _access_f ilt er for LD AP and IdM and ad _access_f ilt er for
AD ) specify which users are granted access to the specified host. The user filter must be used or all
users are denied access. See the examples below:
[domain/example.com]
access_provider = ldap
ldap_access_filter = memberOf=cn=allowedusers,ou=Groups,dc=example,dc=com
[domain/example.com]
access_provider = ad
ad_access_filter = memberOf=cn=allowedusers,ou=Groups,dc=example,dc=com

Note
Offline caching for LD AP access providers is limited to determining whether the user's last
online login attempt was successful. Users that were granted access during their last login will
continue to be granted access while offline.
SSSD can also check results by the authorizedService or host attribute in an entry. In fact, all
options — LD AP filter, authorizedService, and host — can be evaluated, depending on the user entry
and the configuration. The ld ap _access_o rd er parameter lists all access control methods to use,
in order of how they should be evaluated.
[domain/example.com]
access_provider = ldap

267

Deployment G uide

ldap_access_filter = memberOf=cn=allowedusers,ou=Groups,dc=example,dc=com
ldap_access_order = filter, host, authorized_service
The attributes in the user entry to use to evaluate authorized services or allowed hosts can be
customized. Additional access control parameters are listed in the sssd - ld ap ( 5) man page.

12.2.23. Creat ing Domains: Primary Server and Backup Servers
Identity and authentication providers for a domain can be configured for automatic failover. SSSD
attempts to connect to the specified, primary server first. If that server cannot be reached, then SSSD
then goes through the listed backup servers, in order.

Note
SSSD tries to connect to the primary server every 30 seconds, until the connection can be reestablished, and then switches from the backup to the primary.
All of the major service areas have optional settings for primary and backup servers ⁠ [3] .
T ab le 12.11. Primary an d Seco n d ary Server Paramet ers
Service Area

Primary Server At t rib u t e

B acku p Server At t rib u t e

LD AP identity provider
Active D irectory identity
provider
Identity Management (IdM or
IPA) identity provider
Kerberos authentication
provider
Kerberos authentication
provider
Password change provider

ldap_uri
ad_server

ldap_backup_uri
ad_backup_server

ipa_server

ipa_backup_server

krb5_server

krb5_backup_server

krb5_server

krb5_backup_server

ldap_chpass_uri

ldap_chpass_backup_uri

One and only one server can be set as the primary server. (And, optionally, the primary server can be
set to service discovery, using _srv_ rather than a host name.) Multiple backup servers can be set, in
a comma-separate list. The backup server list is in order of preference, so the first server listed is tried
first.
[domain/EXAMPLE]
id_provider = ad
ad_server = ad.example.com
ad_backup_server = ad1.example.com, ad-backup.example.com

12.2.24 . Inst alling SSSD Ut ilit ies
Additional tools to handle the SSSD cache, user entries, and group entries are contained in the
sssd-tools package. This package is not required, but it is useful to install to help administer user
accounts.
~]# yum install sssd-tools

268

⁠Chapt er 1 2 . Configuring Aut hent icat ion

Note
The sssd-tools package is provided by the Optional subscription channel. See Section 7.4.8,
“ Adding the Optional and Supplementary Repositories” for more information on Red Hat
additional channels.

12.2.25. SSSD and UID and GID Numbers
When a user is created — using system tools such as u serad d or through an application such as
Red Hat Identity Management or other client tools — the user is automatically assigned a user ID
number and a group ID number.
When the user logs into a system or service, SSSD caches that user name with the associated
UID /GID numbers. The UID number is then used as the identifying key for the user. If a user with the
same name but a different UID attempts to log into the system, then SSSD treats it as two different
users with a name collision.
What this means is that SSSD does not recognize UID number changes. It interprets it as a different
and new user, not an existing user with a different UID number. If an existing user changes the UID
number, that user is prevented from logging into SSSD and associated services and domains. This
also has an impact on any client applications which use SSSD for identity information; the user with
the conflict will not be found or accessible to those applications.

Important
UID /GID changes are not supported in SSSD .
If a user for some reason has a changed UID /GID number, then the SSSD cache must be cleared for
that user before that user can log in again. For example:
~]# sss_cache -u jsmith
Cleaning the SSSD cache is covered in Section 12.2.28, “ Purging the SSSD Cache” .

12.2.26. Creat ing Local Syst em Users
There can be times when it is useful to seed users into the SSSD database rather than waiting for
users to login and be added.

Note
Adding user accounts manually requires the sssd-tools package to be installed.
When creating new system users, it is possible to create a user within the SSSD local identity
provider domain. This can be useful for creating new system users, for troubleshooting SSSD
configuration, or for creating specialized or nested groups.
New users can be added using the sss_u serad d command.

269

Deployment G uide

At its most basic, the sss_u serad d command only requires the new user name.
~]# sss_useradd jsmith
There are other options (listed in the sss_u serad d ( 8) man page) which can be used to set
attributes on the account, like the UID and GID , the home directory, or groups which the user belongs
to.
~]# sss_useradd --UID 501 --home /home/jsmith --groups admin,dev-group jsmith

12.2.27. Seeding Users int o t he SSSD Cache During Kickst art

Note
Adding user accounts manually requires the sssd-tools package to be installed.
With SSSD , users in a remote domain are not available in a local system until that identity is retrieved
from the identity provider. However, some network interfaces are not available until a user has logged
in — which is not possible if the user identity is somewhere over the network. In that case, it is
possible to seed the SSSD cache with that user identity, associated with the appropriate domain, so
that the user can log in locally and active the appropriate interfaces.
This is done using the sss_seed utility:
sss_seed --domain EXAMPLE.COM --username testuser --password-file /tmp/sssd-pwd.txt
This utility requires options that identify, at a minimum, the user name, domain name, and password.
- - d o main gives the domain name from the SSSD configuration. This domain must already exist
in the SSSD configuration.
- - u sern ame for the short name of the user account.
- - p asswo rd - f ile for the path and name of a file containing a temporary password for the seed
entry. If the user account already exists in the SSSD cache, then the temporary password in this
file overwrites the stored password in the SSSD cache.
Additional account configuration options are listed in the sss_seed ( 8) man page.
This would almost always be run as part of a kickstart or automated setup, so it would be part of a
larger set of scripts, which would also enable SSSD , set up an SSSD domain, and create the
password file. For example:
function make_sssd {
cat <<- _EOF_
[sssd]
domains = LOCAL
services = nss,pam
[nss]
[pam]

270

⁠Chapt er 1 2 . Configuring Aut hent icat ion

[domain/LOCAL]
id_provider = local
auth_provider = local
access_provider = permit
_EOF_
}
make_sssd >> /etc/sssd/sssd.conf
authconfig --enablesssd --enablesssdauth --update
function make_pwdfile {
cat <<1 _EOF_
password
_EOF_
}
make_pwdfile >> /tmp/sssd-pwd.txt
sss_seed --domain EXAMPLE.COM --username testuser --password-file /tmp/sssd-pwd.txt

12.2.28. Managing t he SSSD Cache
SSSD can define multiple domains of the same type and different types of domain. SSSD maintains
a separate database file for each domain, meaning each domain has its own cache. These cache
files are stored in the /var/lib /sss/d b / directory.

Purging t he SSSD Cache
As LD AP updates are made to the identity provider for the domains, it can be necessary to clear the
cache to reload the new information quickly.
The cache purge utility, sss_cach e, invalidates records in the SSSD cache for a user, a domain, or
a group. Invalidating the current records forces the cache to retrieve the updated records from the
identity provider, so changes can be realized quickly.

Note
This utility is included with SSSD in the sssd package.
Most commonly, this is used to clear the cache and update all records:
~]# sss_cache -E
The sss_cach e command can also clear all cached entries for a particular domain:
~]# sss_cache -Ed LD AP1
If the administrator knows that a specific record (user, group, or netgroup) has been updated, then
sss_cach e can purge the records for that specific account and leave the rest of the cache intact:

271

Deployment G uide

~]# sss_cache -u jsmith
T ab le 12.12. C o mmo n sss_cach e O p t io n s
Sh o rt Arg u men t

Lo n g Arg u men t

D escrip t io n

-E

--everything

-d name

--domain name

-G

--groups

-g name

--group name

-N

--netgroups

-n name

--netgroup name

-U

--users

-u name

--user name

Invalidates all cached entries
with the exception of sudo
rules.
Invalidates cache entries for
users, groups, and other
entries only within the specified
domain.
Invalidates all group records. If
- g is also used, - G takes
precedence and - g is ignored.
Invalidates the cache entry for
the specified group.
Invalidates cache entries for all
netgroup cache records. If - n is
also used, - N takes precedence
and - n is ignored.
Invalidates the cache entry for
the specified netgroup.
Invalidates cache entries for all
user records. If the - u option is
also used, - U takes precedence
and - u is ignored.
Invalidates the cache entry for
the specified user.

De le t ing Do m ain Cache File s
All cache files are named for the domain. For example, for a domain named examp leld ap , the cache
file is named cach e_examp leld ap .ld b .
B e caref u l wh en yo u d elet e a cach e f ile. This operation has significant effects:
D eleting the cache file deletes all user data, both identification and cached credentials.
Consequently, do not delete a cache file unless the system is online and can authenticate with a
user name against the domain's servers. Without a credentials cache, offline authentication will
fail.
If the configuration is changed to reference a different identity provider, SSSD will recognize users
from both providers until the cached entries from the original provider time out.
It is possible to avoid this by purging the cache, but the better option is to use a different domain
name for the new provider. When SSSD is restarted, it creates a new cache file with the new name
and the old file is ignored.

12.2.29. Downgrading SSSD

272

⁠Chapt er 1 2 . Configuring Aut hent icat ion

When downgrading — either downgrading the version of SSSD or downgrading the operating
system itself — then the existing SSSD cache needs to be removed. If the cache is not removed, then
SSSD process is dead but a PID file remains. The SSSD logs show that it cannot connect to any of
its associated domains because the cache version is unrecognized.
(Wed Nov 28 21:25:50 2012) [sssd] [sysdb_domain_init_internal] (0x0010): Unknown D B
version [0.14], expected [0.10] for domain AD !
Users are then no longer recognized and are unable to authenticate to domain services and hosts.
After downgrading the SSSD version:
1. D elete the existing cache database files.
~]# rm -rf /var/lib/sss/db/*
2. Restart the SSSD process.
~]# service sssd restart
Stopping sssd:
Starting sssd:

[FAILED ]
[ OK ]

12.2.30. Using NSCD wit h SSSD
SSSD is not designed to be used with the NSCD daemon. Even though SSSD does not directly
conflict with NSCD , using both services can result in unexpected behavior, especially with how long
entries are cached.
The most common evidence of a problem is conflicts with NFS. When using Network Manager to
manage network connections, it may take several minutes for the network interface to come up.
D uring this time, various services attempt to start. If these services start before the network is up and
the D NS servers are available, these services fail to identify the forward or reverse D NS entries they
need. These services will read an incorrect or possibly empty reso lv.co n f file. This file is typically
only read once, and so any changes made to this file are not automatically applied. This can cause
NFS locking to fail on the machine where the NSCD service is running, unless that service is
manually restarted.
To avoid this problem, enable caching for hosts and services in the /et c/n scd .co n f file and rely on
the SSSD cache for the p asswd , g ro u p , and n et g ro u p entries.
Change the /et c/n scd .co n f file:
enable-cache hosts yes
enable-cache passwd no
enable-cache group no
enable-cache netgroup no
With NSCD answering hosts requests, these entries will be cached by NSCD and returned by NSCD
during the boot process. All other entries are handled by SSSD .

12.2.31. T roubleshoot ing SSSD
Section 12.2.31, “ Setting D ebug Logs for SSSD D omains”
Section 12.2.31, “ Checking SSSD Log Files”

273

Deployment G uide

Section 12.2.31, “ Problems with SSSD Configuration”

Se t t ing De bug Lo gs fo r SSSD Do m ains
Each domain sets its own debug log level. Increasing the log level can provide more information
about problems with SSSD or with the domain configuration.
To change the log level, set the debug_level parameter for each section in the sssd .co n f file for
which to produce extra logs. For example:
[domain/LD AP]
cache_credentials = true
d eb u g _level = 9
T ab le 12.13. D eb u g Lo g Levels
Level

D escrip t io n

0

Fatal failures. Anything that would prevent
SSSD from starting up or causes it to cease
running.
Critical failures. An error that doesn't kill the
SSSD , but one that indicates that at least one
major feature is not going to work properly.
Serious failures. An error announcing that a
particular request or operation has failed.
Minor failures. These are the errors that would
percolate down to cause the operation failure of
2.
Configuration settings.
Function data.
Trace messages for operation functions.
Trace messages for internal control functions.
Contents of function-internal variables that may
be interesting.
Extremely low-level tracing information.

1

2
3

4
5
6
7
8
9

Note
In versions of SSSD older than 1.8, debug log levels could be set globally in the [ sssd ]
section. Now, each domain and service must configure its own debug log level.
To copy the global SSSD debug log levels into each configuration area in the SSSD
configuration file, use the sssd _u p d at e_d eb u g _levels.p y script.
python -m SSSD Config.sssd_update_debug_levels.py

Che cking SSSD Lo g File s

274

⁠Chapt er 1 2 . Configuring Aut hent icat ion

SSSD uses a number of log files to report information about its operation, located in the
/var/lo g /sssd / directory. SSSD produces a log file for each domain, as well as an sssd _p am.lo g
and an sssd _n ss.lo g file.
Additionally, the /var/lo g /secu re file logs authentication failures and the reason for the failure.

Pro ble m s wit h SSSD Co nfigurat io n
Q:
SSSD f ails t o st art
A:

SSSD requires that the configuration file be properly set up, with all the required entries, before
the daemon will start.
SSSD requires at least one properly configured domain before the service will start. Without
a domain, attempting to start SSSD returns an error that no domains are configured:
# sssd -d4
[sssd] [ldb] (3): server_sort:Unable to register control with rootdse!
[sssd] [confdb_get_domains] (0): No domains configured, fatal error!
[sssd] [get_monitor_config] (0): No domains configured.
Edit the /et c/sssd /sssd .co n f file and create at least one domain.
SSSD also requires at least one available service provider before it will start. If the problem
is with the service provider configuration, the error message indicates that there are no
services configured:
[sssd] [get_monitor_config] (0): No services configured!
Edit the /et c/sssd /sssd .co n f file and configure at least one service provider.

Important
SSSD requires that service providers be configured as a comma-separated list in a
single services entry in the /et c/sssd /sssd .co n f file. If services are listed in multiple
entries, only the last entry is recognized by SSSD .

Q:
I d o n ' t see an y g ro u p s wit h ' id ' o r g ro u p memb ers wit h ' g et en t g ro u p ' .
A:

This may be due to an incorrect ld ap _sch ema setting in the [ d o main /D O MAIN N AME]
section of sssd .co n f .
SSSD supports RFC 2307 and RFC 2307bis schema types. By default, SSSD uses the more
common RFC 2307 schema.
The difference between RFC 2307 and RFC 2307bis is the way which group membership is
stored in the LD AP server. In an RFC 2307 server, group members are stored as the multivalued memberuid attribute, which contains the name of the users that are members. In an

275

Deployment G uide

RFC2307bis server, group members are stored as the multi-valued member or uniqueMember
attribute which contains the D N of the user or group that is a member of this group. RFC2307bis
allows nested groups to be maintained as well.
If group lookups are not returning any information:
1. Set ld ap _sch ema to rf c2307b is.
2. D elete /var/lib /sss/d b /cach e_D O MAIN N AME.ld b .
3. Restarting SSSD .
If that doesn't work, add this line to sssd .co n f :
ldap_group_name = uniqueMember
Then delete the cache and restart SSSD again.
Q:
Au t h en t icat io n f ails ag ain st LD AP.
A:

To perform authentication, SSSD requires that the communication channel be encrypted. This
means that if sssd .co n f is configured to connect over a standard protocol (ld ap ://), it attempts
to encrypt the communication channel with Start TLS. If sssd .co n f is configured to connect
over a secure protocol (ld ap s://), then SSSD uses SSL.
This means that the LD AP server must be configured to run in SSL or TLS. TLS must be enabled
for the standard LD AP port (389) or SSL enabled on the secure LD APS port (636). With either
SSL or TLS, the LD AP server must also be configured with a valid certificate trust.
An invalid certificate trust is one of the most common issues with authenticating against LD AP. If
the client does not have proper trust of the LD AP server certificate, it is unable to validate the
connection, and SSSD refuses to send the password. The LD AP protocol requires that the
password be sent in plaintext to the LD AP server. Sending the password in plaintext over an
unencrypted connection is a security problem.
If the certificate is not trusted, a syslo g message is written, indicating that TLS encryption could
not be started. The certificate configuration can be tested by checking if the LD AP server is
accessible apart from SSSD . For example, this tests an anonymous bind over a TLS
connection to t est .examp le.co m:
$ ldapsearch -x -Z Z -h test.example.com -b dc=example,dc=com
If the certificate trust is not properly configured, the test fails with this error:
ldap_start_tls: Connect error (-11) additional info: TLS error -8179:Unknown code ___f 13
To trust the certificate:
1. Obtain a copy of the public CA certificate for the certificate authority used to sign the
LD AP server certificate and save it to the local system.
2. Add a line to the sssd .co n f file that points to the CA certificate on the filesystem.
ldap_tls_cacert = /path/to/cacert

276

⁠Chapt er 1 2 . Configuring Aut hent icat ion

3. If the LD AP server uses a self-signed certificate, remove the ld ap _t ls_req cert line from
the sssd .co n f file.
This parameter directs SSSD to trust any certificate issued by the CA certificate, which is
a security risk with a self-signed CA certificate.
Q:
C o n n ect in g t o LD AP servers o n n o n - st an d ard p o rt s f ail.
A:

When running SELinux in enforcing mode, the client's SELinux policy has to be modified to
connect to the LD AP server over the non-standard port. For example:
# semanage port -a -t ldap_port_t -p tcp 1389

Q:
N SS f ails t o ret u rn u ser in f o rmat io n
A:

This usually means that SSSD cannot connect to the NSS service.
Ensure that NSS is running:
# service sssd status
If NSS is running, make sure that the provider is properly configured in the [ n ss] section of
the /et c/sssd /sssd .co n f file. Especially check the filter_users and filter_groups attributes.
Make sure that NSS is included in the list of services that SSSD uses.
Check the configuration in the /et c/n sswit ch .co n f file.

Q:
N SS ret u rn s in co rrect u ser in f o rmat io n
A:

If searches are returning the incorrect user information, check that there are not conflicting user
names in separate domains. When there are multiple domains, set the
use_fully_qualified_domains attribute to t ru e in the /et c/sssd /sssd .co n f file. This
differentiates between different users in different domains with the same name.

Q:
Set t in g t h e p asswo rd f o r t h e lo cal SSSD u ser p ro mp t s t wice f o r t h e p asswo rd
A:

When attempting to change a local SSSD user's password, it may prompt for the password
twice:
[root@clientF11 tmp]# passwd user1000
Changing password for user user1000.
New password:
Retype new password:
New Password:
Reenter new Password:
passwd: all authentication tokens updated successfully.

277

Deployment G uide

This is the result of an incorrect PAM configuration. Ensure that the use_authtok option is
correctly configured in your /et c/p am.d /syst em- au t h file.
Q:
I am t ryin g t o u se su d o ru les wit h an Id en t it y Man ag emen t ( IPA) p ro vid er, b u t n o
su d o ru les are b ein g f o u n d , even t h o u g h su d o is p ro p erly co n f ig u red .
A:

The SSSD client can successfully authenticate to the Identity Management server, and it is
properly searching the LD AP directory for sudo rules. However, it is showing that no rules exist.
For example, in the logs:
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [sdap_sudo_load_sudoers_process]
(0x0400): Receiving sudo rules with base [ou=sudoers,dc=ipa,dc=test]
( T h u Ju n 21 10:37:4 7 2012) [ sssd [ b e[ ip a.t est ] ] ]
[ sd ap _su d o _lo ad _su d o ers_d o n e] ( 0x04 00) : R eceived 0 ru les
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [sdap_sudo_purge_sudoers] (0x0400):
Purging SUD Oers cache of user's [admin] rules
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [sysdb_sudo_purge_byfilter] (0x0400): No
rules matched
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [sysdb_sudo_purge_bysudouser] (0x0400):
No rules matched
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [sdap_sudo_load_sudoers_done] (0x0400):
Sudoers is successfuly stored in cache
(Thu Jun 21 10:37:47 2012) [sssd[be[ipa.test]]] [be_sudo_handler_reply] (0x0200): SUD O
Backend returned: (0, 0, Success)
When using an Identity Management provider for SSSD , SSSD attempts to connect to the
underlying LD AP directory using Kerberos/GSS-API. However, by default, SSSD uses an
anonymous connection to an LD AP server to retrieve sudo rules. This means that SSSD cannot
retrieve the sudo rules from the Identity Management server with its default configuration.
To support retrieving sudo rules with a Kerberos/GSS-API connection, enable GSS-API as the
authentication mechanism in the identity provider configuration in sssd .co n f . For example:
[domain/ipa.example.com]
id_provider = ipa
ipa_server = ipa.example.com
ldap_tls_cacert = /etc/ipa/ca.crt
sudo_provider = ldap
ldap_uri = ldap://ipa.example.com
ldap_sudo_search_base = ou=sudoers,dc=ipa,dc=example,dc=com
ldap_sasl_mech = GSSAPI
ldap_sasl_authid = host/hostname.ipa.example.com
ldap_sasl_realm = IPA.EXAMPLE.COM
krb5_server = ipa.example.com

Q:
Passwo rd lo o ku p s o n larg e d irect o ries can t ake several seco n d s p er req u est . H o w
can t h is b e imp ro ved ?

278

⁠Chapt er 1 2 . Configuring Aut hent icat ion

A:

The initial user lookup is a call to the LD AP server. Unindexed searches are much more
resource-intensive, and therefore take longer, than indexed searches because the server checks
every entry in the directory for a match. To speed up user lookups, index the attributes that are
searched for by SSSD :
uid
uidNumber
gidNumber
gecos

Q:
An Act ive D irect o ry id en t it y p ro vid er is p ro p erly co n f ig u red in my sssd .co n f f ile,
b u t SSSD f ails t o co n n ect t o it , wit h G SS- API erro rs.
A:

SSSD can only connect with an Active D irectory provider using its host name. If the host name
is not given, the SSSD client cannot resolve the IP address to the host, and authentication fails.
For example, with this configuration:
[domain/AD EXAMPLE]
debug_level = 0xFFF0
id_provider = ad
ad _server = 255.255.255.255
ad_domain = example.com
krb5_canonicalize = False
The SSSD client returns this GSS-API failure, and the authentication request fails:
(Fri Jul 27 18:27:44 2012) [sssd[be[AD TEST]]] [sasl_bind_send] (0x0020): ldap_sasl_bind
failed (-2)[Local error]
(Fri Jul 27 18:27:44 2012) [sssd[be[AD TEST]]] [sasl_bind_send] (0x0080): Extended failure
message: [SASL(-1): generic failure: GSSAPI Error: Unspecified GSS failure. Minor code
may provide more information (Cannot determine realm for numeric host address)]
To avoid this error, set the ad_server to the name of the Active D irectory host.

Q:
I co n f ig u red SSSD f o r cen t ral au t h en t icat io n , b u t n o w several o f my ap p licat io n s
( su ch as Firef o x o r Ad o b e) will n o t st art .
A:

Even on 64-bit systems, 32-bit applications require a 32-bit version of SSSD to use to access
the password and identity cache. If a 32-bit version of SSSD is not available, but the system is
configured to use the SSSD cache, then 32-bit applications can fail to start.
For example, Firefox can fail with permission denied errors:
Failed to contact configuration server. See http://www.gnome.org/projects/gconf/
for information. (D etails - 1: IOR file '/tmp/gconfd-somebody/lock/ior'
not opened successfully, no gconfd located: Permission denied 2: IOR
file '/tmp/gconfd-somebody/lock/ior' not opened successfully, no gconfd
located: Permission denied)

279

Deployment G uide

For Adobe Reader, the error shows that the current system user is not recognized:
~]$ acroread
(acroread:12739): GLib-WARNING **: getpwuid_r(): failed due to unknown
user id (366)
Other applications may show similar user or permissions errors.
Q:
SSSD is sh o win g an au t o mo u n t lo cat io n t h at I remo ved .
A:

The SSSD cache for the automount location persists even if the location is subsequently
changed or removed. To update the autofs information in SSSD :
1. Remove the autofs cache, as described in Section 12.2.28, “ Purging the SSSD Cache” .
2. Restart SSSD , as in Section 12.2.3, “ Starting and Stopping SSSD ” .

[3] Mo s t s ervic es d efault to the id entity p ro vid er s erver if a s p ec ific s erver fo r that s ervic e is no t s et.

280

⁠Chapt er 1 3. O penSSH

Chapter 13. OpenSSH
SSH (Secure Shell) is a protocol which facilitates secure communications between two systems
using a client-server architecture and allows users to log into server host systems remotely. Unlike
other remote communication protocols, such as FT P, T eln et , or rlo g in , SSH encrypts the login
session, rendering the connection difficult for intruders to collect unencrypted passwords.
The ssh program is designed to replace older, less secure terminal applications used to log into
remote hosts, such as t eln et or rsh . A related program called scp replaces older programs
designed to copy files between hosts, such as rcp . Because these older applications do not encrypt
passwords transmitted between the client and the server, avoid them whenever possible. Using
secure methods to log into remote systems decreases the risks for both the client system and the
remote host.
Red Hat Enterprise Linux includes the general OpenSSH package, openssh, as well as the OpenSSH
server, openssh-server, and client, openssh-clients, packages. Note, the OpenSSH packages require
the OpenSSL package openssl-libs, which installs several important cryptographic libraries, enabling
OpenSSH to provide encrypted communications.

13.1. T he SSH Prot ocol
13.1.1. Why Use SSH?
Potential intruders have a variety of tools at their disposal enabling them to disrupt, intercept, and reroute network traffic in an effort to gain access to a system. In general terms, these threats can be
categorized as follows:
In t ercep t io n o f co mmu n icat io n b et ween t wo syst ems
The attacker can be somewhere on the network between the communicating parties,
copying any information passed between them. He may intercept and keep the information,
or alter the information and send it on to the intended recipient.
This attack is usually performed using a packet sniffer, a rather common network utility that
captures each packet flowing through the network, and analyzes its content.
Imp erso n at io n o f a p art icu lar h o st
Attacker's system is configured to pose as the intended recipient of a transmission. If this
strategy works, the user's system remains unaware that it is communicating with the wrong
host.
This attack can be performed using a technique known as DNS poisoning, or via so-called IP
spoofing. In the first case, the intruder uses a cracked D NS server to point client systems to a
maliciously duplicated host. In the second case, the intruder sends falsified network
packets that appear to be from a trusted host.
Both techniques intercept potentially sensitive information and, if the interception is made for hostile
reasons, the results can be disastrous. If SSH is used for remote shell login and file copying, these
security threats can be greatly diminished. This is because the SSH client and server use digital
signatures to verify their identity. Additionally, all communication between the client and server
systems is encrypted. Attempts to spoof the identity of either side of a communication does not work,
since each packet is encrypted using a key known only by the local and remote systems.

13.1.2. Main Feat ures

281

Deployment G uide

The SSH protocol provides the following safeguards:
N o o n e can p o se as t h e in t en d ed server
After an initial connection, the client can verify that it is connecting to the same server it had
connected to previously.
N o o n e can cap t u re t h e au t h en t icat io n in f o rmat io n
The client transmits its authentication information to the server using strong, 128-bit
encryption.
N o o n e can in t ercep t t h e co mmu n icat io n
All data sent and received during a session is transferred using 128-bit encryption, making
intercepted transmissions extremely difficult to decrypt and read.
Additionally, it also offers the following options:
It p ro vid es secu re mean s t o u se g rap h ical ap p licat io n s o ver a n et wo rk
Using a technique called X11 forwarding, the client can forward X11 (X Window System)
applications from the server. Note that if you set the Fo rward X11T ru st ed option to yes or
you use SSH with the - Y option, you bypass the X11 SECURITY extension controls, which
can result in a security threat.
It p ro vid es a way t o secu re o t h erwise in secu re p ro t o co ls
The SSH protocol encrypts everything it sends and receives. Using a technique called port
forwarding, an SSH server can become a conduit to securing otherwise insecure protocols,
like POP, and increasing overall system and data security.
It can b e u sed t o creat e a secu re ch an n el
The OpenSSH server and client can be configured to create a tunnel similar to a virtual
private network for traffic between server and client machines.
It su p p o rt s t h e K erb ero s au t h en t icat io n
OpenSSH servers and clients can be configured to authenticate using the GSSAPI (Generic
Security Services Application Program Interface) implementation of the Kerberos network
authentication protocol.

13.1.3. Prot ocol Versions
Two varieties of SSH currently exist: version 1 and version 2. The OpenSSH suite under Red Hat
Enterprise Linux uses SSH version 2, which has an enhanced key exchange algorithm not
vulnerable to the known exploit in version 1. However, for compatibility reasons, the OpenSSH suite
does support version 1 connections as well, although version 1 is disabled by default and needs to
be enabled in the configuration files.

Important
For maximum security, avoid using SSH version 1 and use SSH version 2-compatible servers
and clients whenever possible.

13.1.4 . Event Sequence of an SSH Connect ion

282

⁠Chapt er 1 3. O penSSH

13.1.4 . Event Sequence of an SSH Connect ion
The following series of events help protect the integrity of SSH communication between two hosts.
1. A cryptographic handshake is made so that the client can verify that it is communicating with
the correct server.
2. The transport layer of the connection between the client and remote host is encrypted using a
symmetric cipher.
3. The client authenticates itself to the server.
4. The client interacts with the remote host over the encrypted connection.

1 3.1 .4 .1 . T ranspo rt Laye r
The primary role of the transport layer is to facilitate safe and secure communication between the two
hosts at the time of authentication and during subsequent communication. The transport layer
accomplishes this by handling the encryption and decryption of data, and by providing integrity
protection of data packets as they are sent and received. The transport layer also provides
compression, speeding the transfer of information.
Once an SSH client contacts a server, key information is exchanged so that the two systems can
correctly construct the transport layer. The following steps occur during this exchange:
Keys are exchanged
The public key encryption algorithm is determined
The symmetric encryption algorithm is determined
The message authentication algorithm is determined
The hash algorithm is determined
D uring the key exchange, the server identifies itself to the client with a unique host key. If the client
has never communicated with this particular server before, the server's host key is unknown to the
client and it does not connect. OpenSSH notifies the user that the authenticity of the host cannot be
established and prompts the user to accept or reject it. The user is expected to independently verify
the new host key before accepting it. In subsequent connections, the server's host key is checked
against the saved version on the client, providing confidence that the client is indeed communicating
with the intended server. If, in the future, the host key no longer matches, the user must remove the
client's saved version before a connection can occur.

Warning
Always verify the integrity of a new SSH server. D uring the initial contact, an attacker can
pretend to be the intended SSH server to the local system without being recognized. To verify
the integrity of a new SSH server, contact the server administrator before the first connection or
if a host key mismatch occurs.
SSH is designed to work with almost any kind of public key algorithm or encoding format. After an
initial key exchange creates a hash value used for exchanges and a shared secret value, the two
systems immediately begin calculating new keys and algorithms to protect authentication and future
data sent over the connection.

283

Deployment G uide

After a certain amount of data has been transmitted using a given key and algorithm (the exact
amount depends on the SSH implementation), another key exchange occurs, generating another set
of hash values and a new shared secret value. Even if an attacker is able to determine the hash and
shared secret value, this information is only useful for a limited period of time.

1 3.1 .4 .2 . Aut he nt icat io n
Once the transport layer has constructed a secure tunnel to pass information between the two
systems, the server tells the client the different authentication methods supported, such as using a
private key-encoded signature or typing a password. The client then tries to authenticate itself to the
server using one of these supported methods.
SSH servers and clients can be configured to allow different types of authentication, which gives
each side the optimal amount of control. The server can decide which encryption methods it supports
based on its security model, and the client can choose the order of authentication methods to attempt
from the available options.

1 3.1 .4 .3. Channe ls
After a successful authentication over the SSH transport layer, multiple channels are opened via a
technique called multiplexing ⁠ [4] . Each of these channels handles communication for different
terminal sessions and for forwarded X11 sessions.
Both clients and servers can create a new channel. Each channel is then assigned a different
number on each end of the connection. When the client attempts to open a new channel, the clients
sends the channel number along with the request. This information is stored by the server and is
used to direct communication to that channel. This is done so that different types of sessions do not
affect one another and so that when a given session ends, its channel can be closed without
disrupting the primary SSH connection.
Channels also support flow-control, which allows them to send and receive data in an orderly fashion.
In this way, data is not sent over the channel until the client receives a message that the channel is
open.
The client and server negotiate the characteristics of each channel automatically, depending on the
type of service the client requests and the way the user is connected to the network. This allows great
flexibility in handling different types of remote connections without having to change the basic
infrastructure of the protocol.

13.2. Configuring OpenSSH
13.2.1. Configurat ion Files
There are two different sets of configuration files: those for client programs (that is, ssh , scp , and
sf t p ), and those for the server (the ssh d daemon).
System-wide SSH configuration information is stored in the /et c/ssh / directory as described in
Table 13.1, “ System-wide configuration files” . User-specific SSH configuration information is stored
in ~ /.ssh / within the user's home directory as described in Table 13.2, “ User-specific configuration
files” .
T ab le 13.1. Syst em- wid e co n f ig u rat io n f iles
File

284

D escrip t io n

⁠Chapt er 1 3. O penSSH

File

D escrip t io n

/et c/ssh /mo d u li

Contains D iffie-Hellman groups used for the D iffie-Hellman
key exchange which is critical for constructing a secure
transport layer. When keys are exchanged at the beginning
of an SSH session, a shared, secret value is created which
cannot be determined by either party alone. This value is
then used to provide host authentication.
The default SSH client configuration file. Note that it is
overridden by ~ /.ssh /co n f ig if it exists.
The configuration file for the ssh d daemon.
The D SA private key used by the ssh d daemon.
The D SA public key used by the ssh d daemon.
The RSA private key used by the ssh d daemon for version 1
of the SSH protocol.
The RSA public key used by the ssh d daemon for version 1
of the SSH protocol.
The RSA private key used by the ssh d daemon for version 2
of the SSH protocol.
The RSA public key used by the ssh d daemon for version 2
of the SSH protocol.
The PAM configuration file for the ssh d daemon.
Configuration file for the ssh d service.

/et c/ssh /ssh _co n f ig
/et c/ssh /ssh d _co n f ig
/et c/ssh /ssh _h o st _d sa_key
/et c/ssh /ssh _h o st _d sa_key.p u b
/et c/ssh /ssh _h o st _key
/et c/ssh /ssh _h o st _key.p u b
/et c/ssh /ssh _h o st _rsa_key
/et c/ssh /ssh _h o st _rsa_key.p u b
/et c/p am.d /ssh d
/et c/sysco n f ig /ssh d

T ab le 13.2. U ser- sp ecif ic co n f ig u rat io n f iles
File

D escrip t io n

~ /.ssh /au t h o riz ed _keys

Holds a list of authorized public keys for servers. When the
client connects to a server, the server authenticates the client
by checking its signed public key stored within this file.
Contains the D SA private key of the user.
The D SA public key of the user.
The RSA private key used by ssh for version 2 of the SSH
protocol.
The RSA public key used by ssh for version 2 of the SSH
protocol.
The RSA private key used by ssh for version 1 of the SSH
protocol.
The RSA public key used by ssh for version 1 of the SSH
protocol.
Contains D SA host keys of SSH servers accessed by the
user. This file is very important for ensuring that the SSH
client is connecting the correct SSH server.

~ /.ssh /id _d sa
~ /.ssh /id _d sa.p u b
~ /.ssh /id _rsa
~ /.ssh /id _rsa.p u b
~ /.ssh /id en t it y
~ /.ssh /id en t it y.p u b
~ /.ssh /kn o wn _h o st s

For information concerning various directives that can be used in the SSH configuration files, see the
ssh _co n f ig (5) and ssh d _co n f ig (5) manual pages.

13.2.2. St art ing an OpenSSH Server
In order to run an OpenSSH server, you must have the openssh-server installed (see Section 7.2.4,
“ Installing Packages” for more information on how to install new packages in Red Hat
Enterprise Linux 6).
To start the ssh d daemon, type the following at a shell prompt:

285

Deployment G uide

~]# service ssh d st art
To stop the running ssh d daemon, use the following command:
~]# service ssh d st o p
If you want the daemon to start automatically at the boot time, type:
~]# ch kco n f ig ssh d o n
This will enable the service for levels 2, 3, 4, and 5. For more configuration options, see Chapter 11,
Services and Daemons for the detailed information on how to manage services.
Note that if you reinstall the system, a new set of identification keys will be created. As a result, clients
who had connected to the system with any of the OpenSSH tools before the reinstall will see the
following message:
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@@@@@@@@@@@@@
@ WARNING: REMOTE HOST ID ENTIFICATION HAS CHANGED ! @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS D OING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that the RSA host key has just been changed.
To prevent this, you can backup the relevant files from the /et c/ssh / directory (see Table 13.1,
“ System-wide configuration files” for a complete list), and restore them whenever you reinstall the
system.

13.2.3. Requiring SSH for Remot e Connect ions
For SSH to be truly effective, using insecure connection protocols should be prohibited. Otherwise, a
user's password may be protected using SSH for one session, only to be captured later while
logging in using Telnet. Some services to disable include t eln et , rsh , rlo g in , and vsf t p d .
To disable these services, type the following commands at a shell prompt:
~]# ch kco n f ig
~]# ch kco n f ig
~]# ch kco n f ig
~]# ch kco n f ig

t eln et o f f
rsh o f f
rlo g in o f f
vsf t p d o f f

For more information on runlevels and configuring services in general, see Chapter 11, Services and
Daemons.

13.2.4 . Using Key-Based Aut hent icat ion
To improve the system security even further, you can enforce key-based authentication by disabling
the standard password authentication. To do so, open the /et c/ssh /ssh d _co n f ig configuration file
in a text editor such as vi or n an o , and change the Passwo rd Au t h en t icat io n option as follows:

286

⁠Chapt er 1 3. O penSSH

PasswordAuthentication no
To be able to use ssh , scp , or sf t p to connect to the server from a client machine, generate an
authorization key pair by following the steps below. Note that keys must be generated for each user
separately.
Red Hat Enterprise Linux 6 uses SSH Protocol 2 and RSA keys by default (see Section 13.1.3,
“ Protocol Versions” for more information).

Important
D o not generate key pairs as ro o t , as only ro o t would be able to use those keys.

Note
Before reinstalling your system, back up the ~ /.ssh / directory to keep the generated key pair.
Copy the backed-up data to the home directory in the new system for any user you require,
including root.

1 3.2 .4 .1 . Ge ne rat ing Ke y Pairs
To generate an RSA key pair for version 2 of the SSH protocol, follow these steps:
1. Generate an RSA key pair by typing the following at a shell prompt:
~]$ ssh - keyg en - t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/home/john/.ssh/id_rsa):
2. Press En t er to confirm the default location (that is, ~ /.ssh /id _rsa) for the newly created key.
3. Enter a passphrase, and confirm it by entering it again when prompted to do so. For security
reasons, avoid using the same password as you use to log in to your account.
After this, you will be presented with a message similar to this:
Your identification has been saved in /home/john/.ssh/id_rsa.
Your public key has been saved in /home/john/.ssh/id_rsa.pub.
The key fingerprint is:
e7:97:c7:e2:0e:f9:0e:fc:c4:d7:cb:e5:31:11:92:14 john@penguin.example.com
The key's randomart image is:
+--[ RSA 2048]----+
|
E. |
|
.. |
|
o .|
|
. .|
|
S. .|
|
+ o o ..|

287

Deployment G uide

|
* * +oo|
|
O +..=|
|
o* o.|
+-----------------+
4. Change the permissions of the ~ /.ssh / directory:
~]$ ch mo d 700 ~ /.ssh
5. Copy the content of ~ /.ssh /id _rsa.p u b into the ~ /.ssh /au t h o riz ed _keys on the machine
to which you want to connect, appending it to its end if the file already exists.
6. Change the permissions of the ~ /.ssh /au t h o riz ed _keys file using the following command:
~]$ ch mo d 6 00 ~ /.ssh /au t h o riz ed _keys
To generate a D SA key pair for version 2 of the SSH protocol, follow these steps:
1. Generate a D SA key pair by typing the following at a shell prompt:
~]$ ssh - keyg en - t d sa
Generating public/private dsa key pair.
Enter file in which to save the key (/home/john/.ssh/id_dsa):
2. Press En t er to confirm the default location (that is, ~ /.ssh /id _d sa) for the newly created key.
3. Enter a passphrase, and confirm it by entering it again when prompted to do so. For security
reasons, avoid using the same password as you use to log in to your account.
After this, you will be presented with a message similar to this:
Your identification has been saved in /home/john/.ssh/id_dsa.
Your public key has been saved in /home/john/.ssh/id_dsa.pub.
The key fingerprint is:
81:a1:91:a8:9f:e8:c5:66:0d:54:f5:90:cc:bc:cc:27 john@penguin.example.com
The key's randomart image is:
+--[ D SA 1024]----+
| .oo*o.
|
| ...o Bo
|
| .. . + o.
|
|. . E o
|
| o..o S
|
|. o= .
|
|. +
|
|.
|
|
|
+-----------------+
4. Change the permissions of the ~ /.ssh / directory:
~]$ ch mo d 700 ~ /.ssh
5. Copy the content of ~ /.ssh /id _d sa.p u b into the ~ /.ssh /au t h o riz ed _keys on the machine
to which you want to connect, appending it to its end if the file already exists.

288

⁠Chapt er 1 3. O penSSH

6. Change the permissions of the ~ /.ssh /au t h o riz ed _keys file using the following command:
~]$ ch mo d 6 00 ~ /.ssh /au t h o riz ed _keys
To generate an RSA key pair for version 1 of the SSH protocol, follow these steps:
1. Generate an RSA key pair by typing the following at a shell prompt:
~]$ ssh - keyg en - t rsa1
Generating public/private rsa1 key pair.
Enter file in which to save the key (/home/john/.ssh/identity):
2. Press En t er to confirm the default location (that is, ~ /.ssh /id en t it y) for the newly created
key.
3. Enter a passphrase, and confirm it by entering it again when prompted to do so. For security
reasons, avoid using the same password as you use to log into your account.
After this, you will be presented with a message similar to this:
Your identification has been saved in /home/john/.ssh/identity.
Your public key has been saved in /home/john/.ssh/identity.pub.
The key fingerprint is:
cb:f6:d5:cb:6e:5f:2b:28:ac:17:0c:e4:62:e4:6f:59 john@penguin.example.com
The key's randomart image is:
+--[RSA1 2048]----+
|
|
| ..
|
| oo
|
| +oE
|
| .o S
|
|
=+ . |
|
. = . o . .|
|
. = o o..o|
|
.o o o=o.|
+-----------------+
4. Change the permissions of the ~ /.ssh / directory:
~]$ ch mo d 700 ~ /.ssh
5. Copy the content of ~ /.ssh /id en t it y.p u b into the ~ /.ssh /au t h o riz ed _keys on the machine
to which you want to connect, appending it to its end if the file already exists.
6. Change the permissions of the ~ /.ssh /au t h o riz ed _keys file using the following command:
~]$ ch mo d 6 00 ~ /.ssh /au t h o riz ed _keys
See Section 13.2.4.2, “ Configuring ssh-agent” for information on how to set up your system to
remember the passphrase.

289

Deployment G uide

Important
Never share your private key with anybody; it is for your personal use only.

1 3.2 .4 .2 . Co nfiguring ssh-age nt
To store your passphrase so that you do not have to enter it each time you initiate a connection with
a remote machine, you can use the ssh - ag en t authentication agent. If you are running GNOME, you
can configure it to prompt you for your passphrase whenever you log in and remember it during the
whole session. Otherwise you can store the passphrase for a certain shell prompt.
To save your passphrase during your GNOME session, follow these steps:
1. Make sure you have the openssh-askpass package installed. If not, see Section 7.2.4,
“ Installing Packages” for more information on how to install new packages in Red Hat
Enterprise Linux.
2. Select Syst em → Pref eren ces → St art u p Ap p licat io n s from the panel. The St art u p
Ap p licat io n s Pref eren ces will be started, and the tab containing a list of available startup
programs will be shown by default.

Fig u re 13.1. St art u p Ap p licat io n s Pref eren ces
3. Click the Ad d button on the right, and enter /u sr/b in /ssh - ad d in the C o mman d field.

290

⁠Chapt er 1 3. O penSSH

Fig u re 13.2. Ad d in g n ew ap p licat io n
4. Click Ad d and make sure the checkbox next to the newly added item is selected.

Fig u re 13.3. En ab lin g t h e ap p licat io n
5. Log out and then log back in. A dialog box will appear prompting you for your passphrase.
From this point on, you should not be prompted for a password by ssh , scp , or sf t p .

291

Deployment G uide

Fig u re 13.4 . En t erin g a p assp h rase
To save your passphrase for a certain shell prompt, use the following command:
~]$ ssh - ad d
Enter passphrase for /home/john/.ssh/id_rsa:
Note that when you log out, your passphrase will be forgotten. You must execute the command each
time you log in to a virtual console or a terminal window.

1 3.2 .4 .3. Mult iple re quire d m e t ho ds o f aut he nt icat io n fo r sshd
For higher security, SSH can require multiple methods of authentication to log in successfully, for
example both a passphrase and a public key. Set the R eq u ired Au t h en t icat io n s2 option in the
/et c/ssh /ssh d _co n f ig file as desired, for example by running:
~]# echo " RequiredAuthentications2 publickey,password" >> /etc/ssh/sshd_config
For more information on the available options, see the ssh d _co n f ig ( 5) manual page.

13.3. Using OpenSSH Cert ificat e Aut hent icat ion
13.3.1. Int roduct ion t o SSH Cert ificat es
Using public key cryptography for authentication requires copying the public key from every client to
every server that the client intends to log into. This system does not scale well and can be an
administrative burden. Using a public key from a certificate authority (CA) to authenticate client
certificates removes the need to copy keys between multiple systems. While the X.509 Public Key
Infrastructure Certificate system provides a solution to this issue, there is a submission and
validation process, with associated fees, to go through in order to get a certificate signed. As an
alternative, OpenSSH supports the creation of simple certificates and associated CA infrastructure.
OpenSSH certificates contain a public key, identity information, and validity constraints. They are
signed with a standard SSH public key using the ssh - keyg en utility. The format of the certificate is
described in /u sr/sh are/d o c/o p en ssh - version/PR O T O C O L.cert keys.

292

⁠Chapt er 1 3. O penSSH

The ssh - keyg en utility supports two types of certificates: user and host. User certificates
authenticate users to servers, whereas host certificates authenticate server hosts to users. For
certificates to be used for user or host authentication, ssh d must be configured to trust the CA public
key.

13.3.2. Support for SSH Cert ificat es
Support for certificate authentication of users and hosts using the new OpenSSH certificate format
was introduced in Red Hat Enterprise Linux 6.5, in the openssh-5.3p1-94.el6 package. If required, to
ensure the latest OpenSSH package is installed, enter the following command as ro o t :
~]# yu m in st all o p en ssh
Package openssh-5.3p1-104.el6_6.1.i686 already installed and latest version
Nothing to do

13.3.3. Creat ing SSH CA Cert ificat e Signing Keys
Two types of certificates are required, host certificates and user certificates. It is considered better to
have two separate keys for signing the two certificates, for example ca_u ser_key and ca_h o st _key,
however it is possible to use just one CA key to sign both certificates. It is also easier to follow the
procedures if separate keys are used, so the examples that follow will use separate keys.
The basic format of the command to sign user's public key to create a user certificate is as follows:
ssh-keygen -s ca_user_key -I certificate_ID id_rsa.pub
Where - s indicates the private key used to sign the certificate, - I indicates an identity string, the
certificate_ID, which can be any alpha numeric value. It is stored as a zero terminated string in the
certificate. The certificate_ID is logged whenever the certificate is used for identification and it is also
used when revoking a certificate. Having a long value would make logs hard to read, therefore using
the host name for host certificates and the user name for user certificates is a safe choice.
To sign a host's public key to create a host certificate, add the - h option:
ssh-keygen -s ca_host_key -I certificate_ID -h ssh_host_rsa_key.pub
Host keys are generated on the system by default, to list the keys, enter a command as follows:
~]# ls - l /et c/ssh /ssh _h o st *
-rw-------. 1 root root 668 Jul 9
-rw-r--r--. 1 root root 590 Jul 9
-rw-------. 1 root root 963 Jul 9
-rw-r--r--. 1 root root 627 Jul 9
-rw-------. 1 root root 1671 Jul 9
-rw-r--r--. 1 root root 382 Jul 9

2014 /etc/ssh/ssh_host_dsa_key
2014 /etc/ssh/ssh_host_dsa_key.pub
2014 /etc/ssh/ssh_host_key
2014 /etc/ssh/ssh_host_key.pub
2014 /etc/ssh/ssh_host_rsa_key
2014 /etc/ssh/ssh_host_rsa_key.pub

293

Deployment G uide

Important
It is recommended to create and store CA keys in a safe place just as with any other private
key. In these examples the ro o t user will be used. In a real production environment using an
offline computer with an administrative user account is recommended. For guidance on key
lengths see NIST Special Publication 800-131A.

Pro ced u re 13.1. G en erat in g SSH C A C ert if icat e Sig n in g K eys
1. On the server designated to be the CA, generate two keys for use in signing certificates. These
are the keys that all other hosts need to trust. Choose suitable names, for example
ca_u ser_key and ca_h o st _key. To generate the user certificate signing key, enter the
following command as ro o t :
~]# ssh - keyg en - t rsa - f ~ /.ssh /ca_u ser_key
Generating public/private rsa key pair.
Created directory '/root/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /root/.ssh/ca_user_key.
Your public key has been saved in /root/.ssh/ca_user_key.pub.
The key fingerprint is:
11:14:2f:32:fd:5d:f5:e4:7a:5a:d6:b6:a0:62:c9:1f root@host_name.example.com
The key's randomart image is:
+--[ RSA 2048]----+
|
.+.
o|
|
. o +.|
|
o + . . o|
|
o + . . ..|
|
S . ... *|
|
. . . .*.|
|
= E .. |
|
.o . |
|
. |
+-----------------+
Generate a host certificate signing key, ca_h o st _key, as follows:
~]# ssh - keyg en - t rsa - f ~ /.ssh /ca_h o st _key
Generating public/private rsa key pair.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /root/.ssh/ca_host_key.
Your public key has been saved in /root/.ssh/ca_host_key.pub.
The key fingerprint is:
e4:d5:d1:4f:6b:fd:a2:e3:4e:5a:73:52:91:0b:b7:7a root@host_name.example.com
The key's randomart image is:
+--[ RSA 2048]----+
|
.. |
|
. ....|
|
. . o +oo|
|
o . o *o|
|
S = .|
|
o. .|

294

⁠Chapt er 1 3. O penSSH

|
*.E. |
|
+o= |
|
.oo. |
+-----------------+
If required, confirm the permissions are correct:
~]# ls - la ~ /.ssh
total 40
drwxrwxrwx. 2 root root 4096 May 22 13:18 .
dr-xr-x---. 3 root root 4096 May 8 08:34 ..
-rw-------. 1 root root 1743 May 22 13:15 ca_host_key
-rw-r--r--. 1 root root 420 May 22 13:15 ca_host_key.pub
-rw-------. 1 root root 1743 May 22 13:14 ca_user_key
-rw-r--r--. 1 root root 420 May 22 13:14 ca_user_key.pub
-rw-r--r--. 1 root root 854 May 8 05:55 known_hosts
-r--------. 1 root root 1671 May 6 17:13 ssh_host_rsa
-rw-r--r--. 1 root root 1370 May 7 14:30 ssh_host_rsa-cert.pub
-rw-------. 1 root root 420 May 6 17:13 ssh_host_rsa.pub
2. Create the CA server's own host certificate by signing the server's host public key together
with an identification string such as the host name, the CA server's fully qualified domain name
(FQD N) but without the trailing ., and a validity period. The command takes the following
form:
ssh-keygen -s ~/.ssh/ca_host_key -I certificate_ID -h -Z host_name.example.com -V start:+end /etc/ssh/ssh_host_rsa.pub
The - Z option restricts this certificate to a specific host within the domain. The - V option is for
adding a validity period; this is highly recommend. Where the validity period is intended to be
one year, fifty two weeks, consider the need for time to change the certificates and any holiday
periods around the time of certificate expiry.
For example:
~]# ssh - keyg en - s ~ /.ssh /ca_h o st _key - I h o st _n ame - h - Z
h o st _n ame.examp le.co m - V - 1w:+ 54 w5d /et c/ssh /ssh _h o st _rsa.p u b
Enter passphrase:
Signed host key /root/.ssh/ssh_host_rsa-cert.pub: id " host_name" serial 0 for
host_name.example.com valid from 2015-05-15T13:52:29 to 2016-06-08T13:52:29

13.3.4 . Dist ribut ing and T rust ing SSH CA Public Keys
Hosts that are to allow certificate authenticated log in from users must be configured to trust the CA's
public key that was used to sign the user certificates, in order to authenticate user's certificates. In
this example that is the ca_u ser_key.p u b .
Publish the ca_u ser_key.p u b key and download it to all hosts that are required to allow remote
users to log in. Alternately, copy the CA user public key to all the hosts. In a production environment,
consider copying the public key to an administrator account first. The secure copy command can be
used to copy the public key to remote hosts. The command has the following format:
scp ~/.ssh/ca_user_key.pub root@host_name.example.com:/etc/ssh/

295

Deployment G uide

Where host_name is the host name of a server the is required to authenticate user's certificates
presented during the login process. Ensure you copy the public key not the private key. For example,
as ro o t :
~]# scp ~ /.ssh /ca_u ser_key.p u b ro o t @ h o st _n ame.examp le.co m:/et c/ssh /
The authenticity of host 'host_name.example.com (10.34.74.56)' can't be established.
RSA key fingerprint is fc:23:ad:ae:10:6f:d1:a1:67:ee:b1:d5:37:d4:b0:2f.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'host_name.example.com,10.34.74.56' (RSA) to the list of known
hosts.
root@host_name.example.com's password:
ca_user_key.pub
100% 420 0.4KB/s 00:00
For remote user authentication, CA keys can be marked as trusted per-user in the
~ /.ssh /au t h o riz ed _keys file using the cert - au t h o rit y directive or for global use by means of the
T ru st ed U serC AK eys directive in the /et c/ssh /ssh d _co n f ig file. For remote host authentication,
CA keys can be marked as trusted globally in the /et c/ssh /kn o wn _h o st s file or per-user in the
~ /.ssh /ssh _kn o wn _h o st s file.
Pro ced u re 13.2. T ru st in g t h e U ser Sig n in g K ey
For user certificates which have one or more principles listed, and where the setting is to have
global effect, edit the /et c/ssh /ssh d _co n f ig file as follows:
TrustedUserCAKeys /etc/ssh/ca_user_key.pub
Restart ssh d to make the changes take effect:
~]# service ssh d rest art
To avoid being presented with the warning about an unknown host, a user's system must trust the
CA's public key that was used to sign the host certificates. In this example that is ca_h o st _key.p u b .
Pro ced u re 13.3. T ru st in g t h e H o st Sig n in g K ey
1. Extract the contents of the public key used to sign the host certificate. For example, on the CA:
cat ~ /.ssh /ca_h o st _key.p u b
ssh-rsa AAAAB5Wm.== root@ca-server.example.com
2. To configure client systems to trust servers' signed host certificates, add the contents of the
ca_h o st _key.p u b into the global kn o wn _h o st s file. This will automatically check a
server's host advertised certificate against the CA public key for all users every time a new
machine is connected to in the domain *.examp le.co m. Login as ro o t and configure the
/et c/ssh /ssh _kn o wn _h o st s file, as follows:
~]# vi /et c/ssh /ssh _kn o wn _h o st s
# A CA key, accepted for any host in *.example.com
@cert-authority *.example.com ssh-rsa AAAAB5Wm.
Where ssh - rsa AAAAB5Wm. is the contents of ca_h o st _key.p u b . The above configures the
system to trust the CA servers host public key. This enables global authentication of the
certificates presented by hosts to remote users.

296

⁠Chapt er 1 3. O penSSH

13.3.5. Creat ing SSH Cert ificat es
A certifcate is a signed public key. The user's and host's public keys must be copied to the CA server
for signing by the CA server's private key.

Important
Copying many keys to the CA to be signed can create confusion if they are not uniquely
named. If the default name is always used then the latest key to be copied will overwrite the
previously copied key, which may be an acceptable method for one administrator. In the
example below the default name is used. In a production environment, consider using easily
recognizable names. It is recommend to have a designated directory on the CA server owned
by an administrative user for the keys to be copied into. Copying these keys to the ro o t user's
/et c/ssh / directory is not recommend. In the examples below an account named ad min with a
directory named keys/ will be used.
Create an administrator account, in this example ad min , and a directory to receive the user's keys.
For example:
~]$ mkd ir keys
Set the permissions to allow keys to be copied in:
~]$ ch mo d o + w keys
ls - la keys
total 8
drwxrwxrwx. 2 admin admin 4096 May 22 16:17 .
drwx------. 3 admin admin 4096 May 22 16:17 ..

1 3.3.5 .1 . Cre at ing SSH Ce rt ificat e s t o Aut he nt icat e Ho st s
The command to sign a host certificate has the following format:
ssh-keygen -s ca_host_key -I host_name -h ssh_host_rsa_key.pub
The host certificate will named ssh _h o st _rsa_key- cert .p u b .
Pro ced u re 13.4 . G en erat in g a H o st C ert if icat e
To authenticate a host to a user, a public key must be generated on the host, passed to the CA
server, signed by the CA, and then passed back to be stored on the host to present to a user
attempting to log into the host.
1. Host keys are generated automatically on the system. To list them enter the following
command:
~]# ls - l /et c/ssh /ssh _h o st *
-rw-------. 1 root root 668 May 6 14:38 /etc/ssh/ssh_host_dsa_key
-rw-r--r--. 1 root root 590 May 6 14:38 /etc/ssh/ssh_host_dsa_key.pub
-rw-------. 1 root root 963 May 6 14:38 /etc/ssh/ssh_host_key

297

Deployment G uide

-rw-r--r--. 1 root root 627 May 6 14:38 /etc/ssh/ssh_host_key.pub
-rw-------. 1 root root 1679 May 6 14:38 /etc/ssh/ssh_host_rsa_key
-rw-r--r--. 1 root root 382 May 6 14:38 /etc/ssh/ssh_host_rsa_key.pub
2. Copy the chosen public key to the server designated as the CA. For example, from the host:
~]# scp /et c/ssh /ssh _h o st _rsa_key.p u b ad min @ caserver.examp le.co m:~ /keys/ssh _h o st _rsa_key.p u b
The authenticity of host 'ca-server.example.com (10.34.74.58)' can't be established.
RSA key fingerprint is b0:e5:ea:b8:75:e2:f0:b1:fe:5b:07:39:7f:58:64:d9.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'ca-server.example.com,10.34.74.58' (RSA) to the list of
known hosts.
admin@ca-server.example.com's password:
ssh_host_rsa_key.pub
100% 382 0.4KB/s 00:00
Alternately, from the CA:
~]$ scp ro o t @ h o st _n ame.examp le.co m:/et c/ssh /ssh _h o st _rsa_key.p u b
~ /keys/ssh _h o st _rsa_key.p u b
3. On the CA server, sign the host's public key. For example, as ro o t :
~]# ssh - keyg en - s ~ /.ssh /ca_h o st _key - I h o st _n ame - h - Z
h o st _n ame.examp le.co m - V - 1d :+ 54 w
/h o me/ad min /keys/ssh _h o st _rsa_key.p u b
Enter passphrase:
Signed host key /home/admin/keys/ssh_host_rsa_key-cert.pub: id " host_name" serial 0
for host_name.example.com valid from 2015-05-26T12:21:54 to 2016-06-08T12:21:54
Where host_name is the host name of the system requiring the certificate.
4. Copy the certificate to the host. For example, from the CA:
~]# scp /h o me/ad min /keys/ssh _h o st _rsa_key- cert .p u b
ro o t @ h o st _n ame.examp le.co m:/et c/ssh /
root@host_name.example.com's password:
ssh_host_rsa_key-cert.pub
100% 1384 1.5KB/s 00:00
5. Configure the host to present the certificate to a user's system when a user initiates the login
process. As ro o t , edit the /et c/ssh /ssh d _co n f ig file as follows:
HostCertificate /etc/ssh/ssh_host_rsa_key-cert.pub
6. Restart ssh d to make the changes take effect:
~]# service ssh d rest art
7. On user's systems. remove keys belonging to hosts from the ~ /.ssh /kn o wn _h o st s file if the
user has previously logged into the host configured above. When a user logs into the host
they should no longer be presented with the warning about the hosts authenticity.
To test the host certificate, on a client system, ensure the client has set up the global
/et c/ssh /kn o wn _h o st s file, as described in Procedure 13.3, “ Trusting the Host Signing Key” , and

298

⁠Chapt er 1 3. O penSSH

that the server's public key is not in the ~ /.ssh /kn o wn _h o st s file. Then attempt to log into the server
over SSH as a remote user. You should not see a warning about the authenticity of the host. If
required, add the - v option to the SSH command to see logging information.

1 3.3.5 .2 . Cre at ing SSH Ce rt ificat e s fo r Aut he nt icat ing Use rs
To sign a user's certificate, use a command in the following format:
ssh-keygen -s ca_user_key -I user_name -Z user_name -V -start:+end id_rsa.pub
The resulting certificate will be named id _rsa- cert .p u b .
The default behavior of OpenSSH is that a user is allowed to log in as a remote user if one of the
principals specified in the certificate matches the remote user's name. This can be adjusted in the
following ways:
Add more user's names to the certificate during the signing process using the - Z option:
-Z " name1[,name2,...]"
On the user's system, add the public key of the CA in the ~ /.ssh /au t h o riz ed _keys file using the
cert - au t h o rit y directive and list the principals names as follows:
~]# vi ~ /.ssh /au t h o riz ed _keys
# A CA key, accepted for any host in *.example.com
@cert-authority principals=" name1,name2" *.example.com ssh-rsa AAAAB5Wm.
On the server, create an Au t h o riz ed Prin cip alsFile file, either per user or glabally, and add the
principles' names to the file for those users allowed to log in. Then in the /et c/ssh /ssh d _co n f ig
file, specify the file using the Au t h o riz ed Prin cip alsFile directive.
Pro ced u re 13.5. G en erat in g a U ser C ert if icat e
To authenticate a user to a remote host, a public key must be generated by the user, passed to the CA
server, signed by the CA, and then passed back to be stored by the user for use when logging in to a
host.
1. On client systems, login as the user who requires the certificate. Check for available keys as
follows:
~]$ ls - l ~ /.ssh /
If no suitable public key exists, generate one and set the directory permissions if the directory
is not the default directory. For example, enter the following command:
~]$ ssh - keyg en - t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/home/user1/.ssh/id_rsa):
Created directory '/home/user1/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/user1/.ssh/id_rsa.
Your public key has been saved in /home/user1/.ssh/id_rsa.pub.
The key fingerprint is:
b1:f8:26:a7:46:87:c3:60:54:a3:6d:85:0d:60:fe:ce user1@host1.example.com

299

Deployment G uide

The key's randomart image is:
+--[ RSA 2048]----+
| oo++.
|
| o.o.o.
|
| .o o .
|
| oo . o
|
| . oo.S
|
| o=..
|
| .Eo+
|
|
.=
|
| ..
|
+-----------------+
By default the directory permissions for a user's keys are d rwx- - - - - - ., or octal 0700. If
required, confirm the permissions are correct:
~]$ ls - la ~ /.ssh
total 16
drwx------. 2 user1 user1 4096 May 7 12:37 .
drwx------. 3 user1 user1 4096 May 7 12:37 ..
-rw-------. 1 user1 user1 1679 May 7 12:37 id_rsa
-rw-r--r--. 1 user1 user1 421 May 7 12:37 id_rsa.pub
See Section 13.2.4, “ Using Key-Based Authentication” for more examples of key generation
and for instructions on setting the correct directory permissions.
2. The chosen public key must be copied to the server designated as the CA, in order to be
signed. The secure copy command can be used to do this, the command has the following
format:
scp ~/.ssh/id_protocol.pub admin@ca_server.example.com:~/keys/
Where protocol is the part of the file name indicating the protocol used to generate the key, for
example rsa, admin is an account on the CA server, and /keys/ is a directory setup to receive
the keys to be signed.
Copy the chosen public key to the server designated as the CA. For example:
~]$ scp ~ /.ssh /id _rsa.p u b ad min @ ca- server.examp le.co m:~ /keys/
admin@ca-server.example.com's password:
id_rsa.pub
100% 421 0.4KB/s 00:00
If you have configured the client system to trust the host signing key as described in
Procedure 13.3, “ Trusting the Host Signing Key” then you should not see a warning about
the authenticity of the remote host.
3. On the CA server, sign the user's public key. For example, as ro o t :
~]# ssh - keyg en - s ~ /.ssh /ca_u ser_key - I u ser1 - Z u ser1 - V - 1d :+ 54 w
/h o me/ad min /keys/id _rsa.p u b
Enter passphrase:
Signed user key /home/admin/keys/id_rsa-cert.pub: id " user1" serial 0 for
host_name.example.com valid from 2015-05-21T16:43:17 to 2016-06-03T16:43:17
4. Copy the resulting certificate to the user's ~ /.ssh / directory on their system. For example:

300

⁠Chapt er 1 3. O penSSH

~]# scp /h o me/ad min /keys/id _rsa- cert .p u b
u ser1@ h o st _n ame.examp le.co m:~ /.ssh /
user1@host_name.example.com's password:
id_rsa-cert.pub
100% 1498 1.5KB/s 00:00
5. If using the standard file names and location then no further configuration is required as the
SSH daemon will search for user certificates ending in - cert .p u b and use them
automatically if it finds them. Note that the default location and file names for for SSH version
2 keys are: ~ /.ssh /id _d sa, ~ /.ssh /id _ecd sa and ~ /.ssh /id _rsa as explained in the
ssh _co n f ig ( 5) manual page. If you use these locations and naming conventions then there
is no need for editing the configuration files to enable ssh d to present the certificate. They
will be used automatically when logging in to a remote system. In this is the case then skip to
step 6.
If required to use a non-default directory or file naming convention, then as ro o t , add the
following line to the /et c/ssh /ssh _co n f ig or ~ /.ssh /co n f ig files:
IdentityFile ~/path/key_file
Note that this must be the private key name, do not had .p u b or - cert .p u b . Ensure the file
permission are correct. For example:
~]$ ls - la ~ /.ssh /co n f ig
-rw-rw-r--. 1 user1 user1 36 May 27 21:49 /home/user1/.ssh/config
ch mo d 700 ~ /.ssh /co n f ig
~]$ ls - la ~ /.ssh /co n f ig
-rwx------. 1 user1 user1 36 May 27 21:49 /home/user1/.ssh/config
This will enable the user of this system to be authenticated by a user certificate when logging
into a remote system configured to trust the CA user certificate signing key.
6. To test the user certificate, attempt to log into a server over SSH from the user's account. You
should do this as the user listed as a principle in the certificate, if any are specified. You
should not be prompted for a password. If required, add the - v option to the SSH command to
see logging information.

13.3.6. Signing an SSH Cert ificat e Using a PKCS#11 T oken
It is possible to sign a host key using a CA key stored in a PKCS#11 token by providing the token
library using the - D and identifying the CA key by providing its public half as an argument to the - s
option:
ssh-keygen -s ca_host_key.pub -D libpkcs11.so -I certificate_ID host_key.pub
In all cases, certificate_ID is a “ key identifier” that is logged by the server when the certificate is used
for authentication.
Certificates may be configured to be valid only for a set of users or host names, the principals. By
default, generated certificates are valid for all users or hosts. To generate a certificate for a specified
set of principals, use a comma separated list with the - Z option as follows:
ssh-keygen -s ca_user_key.pub -D libpkcs11.so -I certificate_ID -Z user1,user2 id_rsa.pub
and for hosts:

301

Deployment G uide

ssh-keygen -s ca_host_key.pub -D libpkcs11.so -I certificate_ID -h -Z host.domain
ssh_host_rsa_key.pub
Additional limitations on the validity and use of user certificates may be specified through certificate
options. A certificate option may disable features of the SSH session, may be valid only when
presented from particular source addresses or may force the use of a specific command. For a list of
valid certificate options, see the ssh - keyg en ( 1) manual page for the - O option.
Certificates may be defined to be valid for a specific lifetime. The - V option allows specifying a
certificates start and end times. For example:
ssh-keygen -s ca_user_key -I certificate_ID id_rsa.pub -V " -1w:+54w5d"
A certificate that is presented at a time outside this range will not be considered valid. By default,
certificates are valid indefinitely starting from UNIX Epoch.

13.3.7. Viewing an SSH CA Cert ificat e
To view a certificate, use the - L to list the contents. For example, for a user's certificate:
~]$ ssh - keyg en - L - f ~ /.ssh /id _rsa- cert .p u b
/home/user1/.ssh/id_rsa-cert.pub:
Type: ssh-rsa-cert-v01@openssh.com user certificate
Public key: RSA-CERT 3c:9d:42:ed:65:b6:0f:18:bf:52:77:c6:02:0e:e5:86
Signing CA: RSA b1:8e:0b:ce:fe:1b:67:59:f1:74:cd:32:af:5f:c6:e8
Key ID : " user1"
Serial: 0
Valid: from 2015-05-27T00:09:16 to 2016-06-09T00:09:16
Principals:
user1
Critical Options: (none)
Extensions:
permit-X11-forwarding
permit-agent-forwarding
permit-port-forwarding
permit-pty
permit-user-rc
To vew a host certificate:
~]# ssh - keyg en - L - f /et c/ssh /ssh _h o st _rsa_key- cert .p u b
/etc/ssh/ssh_host_rsa_key-cert.pub:
Type: ssh-rsa-cert-v01@openssh.com host certificate
Public key: RSA-CERT 1d:71:61:50:05:9b:ec:64:34:27:a5:cc:67:24:03:23
Signing CA: RSA e4:d5:d1:4f:6b:fd:a2:e3:4e:5a:73:52:91:0b:b7:7a
Key ID : " host_name"
Serial: 0
Valid: from 2015-05-26T17:19:01 to 2016-06-08T17:19:01
Principals:
host_name.example.com
Critical Options: (none)
Extensions: (none)

13.3.8. Revoking an SSH CA Cert ificat e

302

⁠Chapt er 1 3. O penSSH

13.3.8. Revoking an SSH CA Cert ificat e
If a certificate is stolen, it should be revoked. Although OpenSSH does not provide a mechanism to
distribute the revocation list it is still easier to create the revocation list and distribute it by other
means then to change the CA keys and all host and user certificates previously created and
distributed.
Keys can be revoked by adding them to the revo ked _keys file and specifying the file name in the
ssh d _co n f ig file as follows:
RevokedKeys /etc/ssh/revoked_keys
Note that if this file is not readable, then public key authentication will be refused for all users.
To test if a key has been revoked, query the revocation list for the presence of the key. Use a
command as follows:
ssh-keygen -Qf /etc/ssh/revoked_keys ~/.ssh/id_rsa.pub
A user can revoke a CA certificate by changing the cert - au t h o rit y directive to revo ke in the
kn o wn _h o st s file.

13.4 . OpenSSH Client s
To connect to an OpenSSH server from a client machine, you must have the openssh-clients and
openssh packages installed (see Section 7.2.4, “ Installing Packages” for more information on how to
install new packages in Red Hat Enterprise Linux).

13.4 .1. Using t he ssh Ut ilit y
The ssh utility allows you to log in to a remote machine and execute commands there. It is a secure
replacement for the rlo g in , rsh , and t eln et programs.
Similarly to the t eln et command, log in to a remote machine by using the following command:
ssh hostname
For example, to log in to a remote machine named p en g u in .examp le.co m, type the following at a
shell prompt:
~]$ ssh p en g u in .examp le.co m
This will log you in with the same user name you are using on the local machine. If you want to
specify a different user name, use a command in the following form:
ssh username@hostname
For example, to log in to p en g u in .examp le.co m as jo h n , type:
~]$ ssh jo h n @ p en g u in .examp le.co m
The first time you initiate a connection, you will be presented with a message similar to this:

303

Deployment G uide

The authenticity of host 'penguin.example.com' can't be established.
RSA key fingerprint is 94:68:3a:3a:bc:f3:9a:9b:01:5d:b3:07:38:e2:11:0c.
Are you sure you want to continue connecting (yes/no)?
Type yes to confirm. You will see a notice that the server has been added to the list of known hosts,
and a prompt asking for your password:
Warning: Permanently added 'penguin.example.com' (RSA) to the list of known hosts.
john@penguin.example.com's password:

Important
Update the host key of an SSH server if the key changes. The client notifies the user that the
connection cannot proceed until the server's host key is deleted from the
~ /.ssh /kn o wn _h o st s file. Contact the system administrator of the SSH server to verify the
server is not compromised, then remove the line with the name of the remote machine at the
beginning.
After entering the password, you will be provided with a shell prompt for the remote machine.
Alternatively, the ssh program can be used to execute a command on the remote machine without
logging in to a shell prompt:
ssh [username@]hostname command
For example, the /et c/red h at - release file provides information about the Red Hat Enterprise Linux
version. To view the contents of this file on p en g u in .examp le.co m, type:
~]$ ssh jo h n @ p en g u in .examp le.co m cat /et c/red h at - release
john@penguin.example.com's password:
Red Hat Enterprise Linux Server release 6.2 (Santiago)
After you enter the correct password, the user name will be displayed, and you will return to your
local shell prompt.

13.4 .2. Using t he scp Ut ilit y
scp can be used to transfer files between machines over a secure, encrypted connection. In its
design, it is very similar to rcp .
To transfer a local file to a remote system, use a command in the following form:
scp localfile username@ hostname:remotefile
For example, if you want to transfer t ag list .vim to a remote machine named
p en g u in .examp le.co m, type the following at a shell prompt:
~]$ scp t ag list .vim jo h n @ p en g u in .examp le.co m:.vim/p lu g in /t ag list .vim
john@penguin.example.com's password:
taglist.vim
100% 144KB 144.5KB/s 00:00

304

⁠Chapt er 1 3. O penSSH

Multiple files can be specified at once. To transfer the contents of .vim/p lu g in / to the same directory
on the remote machine p en g u in .examp le.co m, type the following command:
~]$ scp .vim/p lu g in /* jo h n @ p en g u in .examp le.co m:.vim/p lu g in /
john@penguin.example.com's password:
closetag.vim
100% 13KB 12.6KB/s 00:00
snippetsEmu.vim
100% 33KB 33.1KB/s 00:00
taglist.vim
100% 144KB 144.5KB/s 00:00
To transfer a remote file to the local system, use the following syntax:
scp username@ hostname:remotefile localfile
For instance, to download the .vimrc configuration file from the remote machine, type:
~]$ scp jo h n @ p en g u in .examp le.co m:.vimrc .vimrc
john@penguin.example.com's password:
.vimrc
100% 2233 2.2KB/s 00:00

13.4 .3. Using t he sf t p Ut ilit y
The sf t p utility can be used to open a secure, interactive FTP session. In its design, it is similar to
f t p except that it uses a secure, encrypted connection.
To connect to a remote system, use a command in the following form:
sf t p username@ hostname
For example, to log in to a remote machine named p en g u in .examp le.co m with jo h n as a user
name, type:
~]$ sf t p jo h n @ p en g u in .examp le.co m
john@penguin.example.com's password:
Connected to penguin.example.com.
sftp>
After you enter the correct password, you will be presented with a prompt. The sf t p utility accepts a
set of commands similar to those used by f t p (see Table 13.3, “ A selection of available sftp
commands” ).
T ab le 13.3. A select io n o f availab le sf t p co mman d s
C o mman d

D escrip t io n

ls [directory]

List the content of a remote directory. If none is supplied, a
current working directory is used by default.
Change the remote working directory to directory.
Create a remote directory.
Remove a remote directory.
Transfer localfile to a remote machine.
Transfer remotefile from a remote machine.

cd directory
mkd ir directory
rmd ir path
p u t localfile [remotefile]
g et remotefile [localfile]

For a complete list of available commands, see the sf t p (1) manual page.

305

Deployment G uide

13.5. More T han a Secure Shell
A secure command-line interface is just the beginning of the many ways SSH can be used. Given the
proper amount of bandwidth, X11 sessions can be directed over an SSH channel. Or, by using
TCP/IP forwarding, previously insecure port connections between systems can be mapped to specific
SSH channels.

13.5.1. X11 Forwarding
To open an X11 session over an SSH connection, use a command in the following form:
ssh - Y username@ hostname
For example, to log in to a remote machine named p en g u in .examp le.co m with jo h n as a user
name, type:
~]$ ssh - Y jo h n @ p en g u in .examp le.co m
john@penguin.example.com's password:
When an X program is run from the secure shell prompt, the SSH client and server create a new
secure channel, and the X program data is sent over that channel to the client machine transparently.
X11 forwarding can be very useful. For example, X11 forwarding can be used to create a secure,
interactive session of the Prin t er C o n f ig u rat io n utility. To do this, connect to the server using ssh
and type:
~]$ syst em- co n f ig - p rin t er &
The Prin t er C o n f ig u rat io n T o o l will appear, allowing the remote user to safely configure printing
on the remote system.
Please note that X11 Forwarding does not distinguish between trusted and untrusted forwarding.

13.5.2. Port Forwarding
SSH can secure otherwise insecure T C P/IP protocols via port forwarding. When using this
technique, the SSH server becomes an encrypted conduit to the SSH client.
Port forwarding works by mapping a local port on the client to a remote port on the server. SSH can
map any port from the server to any port on the client. Port numbers do not need to match for this
technique to work.

Note
If you want to use reserved port numbers, please note that setting up port forwarding to listen
on ports below 1024 requires ro o t level access.
To create a TCP/IP port forwarding channel which listens for connections on the lo calh o st , use a
command in the following form:
ssh - L local-port:remote-hostname:remote-port username@ hostname

306

⁠Chapt er 1 3. O penSSH

For example, to check email on a server called mail.examp le.co m using PO P3 through an
encrypted connection, use the following command:
~]$ ssh - L 1100:mail.examp le.co m:110 mail.examp le.co m
Once the port forwarding channel is in place between the client machine and the mail server, direct a
POP3 mail client to use port 1100 on the lo calh o st to check for new email. Any requests sent to port
1100 on the client system will be directed securely to the mail.examp le.co m server.
If mail.examp le.co m is not running an SSH server, but another machine on the same network is,
SSH can still be used to secure part of the connection. However, a slightly different command is
necessary:
~]$ ssh - L 1100:mail.examp le.co m:110 o t h er.examp le.co m
In this example, POP3 requests from port 1100 on the client machine are forwarded through the SSH
connection on port 22 to the SSH server, o t h er.examp le.co m. Then, o t h er.examp le.co m
connects to port 110 on mail.examp le.co m to check for new email. Note that when using this
technique, only the connection between the client system and o t h er.examp le.co m SSH server is
secure.
Port forwarding can also be used to get information securely through network firewalls. If the firewall
is configured to allow SSH traffic via its standard port (that is, port 22) but blocks access to other
ports, a connection between two hosts using the blocked ports is still possible by redirecting their
communication over an established SSH connection.

Important
The connection is only as secure as the client system because forwarding connections in this
way allows any user on the client system to connect to that service. If the client system
becomes compromised, an attacker can also access the forwarded services.
If preferred, disable this functionality on the server by specifying a N o parameter for the
Allo wT cp Fo rward in g line in the /et c/ssh /ssh d _co n f ig file and restarting the ssh d
service.

13.6. Addit ional Resources
For more information about OpenSSH and OpenSSL, see the resources listed below.

13.6.1. Inst alled Document at ion
ssh d (8) — a manual page for the ssh d daemon.
ssh (1) — a manual page for the ssh client.
scp (1) — a manual page for the scp utility.
sf t p (1) — a manual page for the sf t p utility.
ssh - keyg en (1) — a manual page for the ssh - keyg en utility.
ssh _co n f ig (5) — a manual page with a full description of available SSH client configuration

307

Deployment G uide

options.
ssh d _co n f ig (5) — a manual page with a full description of available SSH daemon
configuration options.
/u sr/sh are/d o c/o p en ssh - version/ Contains detailed information on the protocols supported by
OpenSSH.

13.6.2. Useful Websit es
h t t p ://www.o p en ssh .co m/
The OpenSSH home page containing further documentation, frequently asked questions,
links to the mailing lists, bug reports, and other useful resources.
h t t p ://www.o p en ssl.o rg /
The OpenSSL home page containing further documentation, frequently asked questions,
links to the mailing lists, and other useful resources.

[4] A multip lexed c o nnec tio n c o ns is ts o f s everal s ig nals b eing s ent o ver a s hared , c o mmo n med ium.
With SSH, d ifferent c hannels are s ent o ver a c o mmo n s ec ure c o nnec tio n.

308

⁠Chapt er 1 4 . T igerVNC

Chapter 14. TigerVNC
T ig erVN C (Tiger Virtual Network Computing) is a system for graphical desktop sharing which allows
you to remotely control other computers.
T ig erVN C works on the client-server principle: a server shares its output (vn cserver) and a clien t
(vn cviewer) connects to the server.

14 .1. VNC Server
vn cserver is a utility which starts a VNC (Virtual Network Computing) desktop. It runs Xvn c with
appropriate options and starts a window manager on the VNC desktop. vn cserver allows users to
run separate sessions in parallel on a machine which can then be accessed by any number of
clients from anywhere.

14 .1.1. Inst alling VNC Server
To install the T ig erVN C server, run the following command as ro o t :
~]# yu m in st all t ig ervn c- server

14 .1.2. Configuring VNC Server
The VNC server can be configured to start a display for one or more users, provided that accounts for
the users exist on the system, with optional parameters such as for display settings, network address
and port, and security settings.
Pro ced u re 14 .1. C o n f ig u rin g a VN C D isp lay f o r a Sin g le U ser
Specify the user name and the display number by editing /et c/sysco n f ig /vn cservers and
adding a line in the following format:
VNCSERVERS=" display_number:user"
The VNC user names must correspond to users of the system.

Examp le 14 .1. Set t in g t h e D isp lay N u mb er f o r a U ser
For example, to configure display number 3 for user jo e, open the configuration file for editing:
~]# vi /et c/sysco n f ig /vn cservers
Add a line as follows:
VNCSERVERS=" 3:joe"
Save and close the file.
In the example above, display number 3 and the user jo e are set. D o not use 0 as the display
number since the main X display of a workstation is usually indicated as 0.

309

Deployment G uide

Pro ced u re 14 .2. C o n f ig u rin g a VN C D isp lay f o r Mu lt ip le U sers
To set a VNC display for more than one user, specify the user names and display numbers by
editing /et c/sysco n f ig /vn cservers and adding a line in the following format:
VNCSERVERS=" display_number:user display_number:user"
The VNC user names must correspond to users of the system.

Examp le 14 .2. Set t in g t h e D isp lay N u mb ers f o r T wo U sers
For example, to configure two users, open the configuration file for editing:
~]# vi /et c/sysco n f ig /vn cservers
Add a line as follows:
VNCSERVERS=" 3:joe 4:jill"

Pro ced u re 14 .3. C o n f ig u rin g VN C D isp lay Arg u men t s
Specify additional settings in the /et c/sysco n f ig /vn cservers file by adding arguments using the
VNCSERVERARGS directive as follows:
VNCSERVERS=" display_number:user display_number:user"
VNCSERVERARGS[display_number]=" arguments"
T ab le 14 .1. Freq u en t ly U sed VN C Server Paramet ers
VN C SER VER AR G S

D ef in it io n

-geometry

specifies the size of the VNC desktop to be created, default is
1024x768.
prevents connections to your VNC server through TCP (Transmission
Control Protocol)
prevents remote VNC clients from connecting except when doing so
through a secure tunnel

-nolisten tcp
-localhost

See the Xvn c( 1) man page for further options.

Examp le 14 .3. Set t in g vn cserver Arg u men t s
Following on from the example above, to add arguments for two users, edit the
/et c/sysco n f ig /vn cservers file as follows:
VNCSERVERS=" 3:joe 4:jill"
VNCSERVERARGS[1]=" -geometry 800x600 -nolisten tcp -localhost"
VNCSERVERARGS[2]=" -geometry 1920×1080 -nolisten tcp -localhost"

Pro ced u re 14 .4 . C o n f ig u rin g VN C U ser Passwo rd s

310

⁠Chapt er 1 4 . T igerVNC

To set the VNC password for all users defined in the /et c/sysco n f ig /vn cservers file, enter the
following command as ro o t :
~]# vn cp asswd
Password:
Verify:
To set the VNC password individually for a user:
~]# su - user
~]$ vn cp asswd
Password:
Verify:

Important
The stored password is not encrypted; anyone who has access to the password file can
find the plain-text password.

14 .1.3. St art ing VNC Server
In order to start a VNC desktop, the vn cserver utility is used. It is a Perl script which simplifies the
process of starting an Xvnc server. It runs Xvnc with appropriate options and starts a window
manager on the VNC desktop. There are three ways to start vn cserver:
You can allow vn cserver to choose the first available display number, start Xvnc with that
display number, and start the default window manager in the Xvnc session. All these steps are
provided by one command:
~]$ vn cserver
You will be prompted to enter a VNC password the first time the command is run if no VNC
password has been set.
Alternately, you can specify a specific display number:
vncserver :display_number
vn cserver attempts to start Xvnc with that display number and exits if the display number is not
available.
For example:
~]$ vn cserver :20
Alternately, to start VNC server with displays for the users configured in the
/et c/sysco n f ig /vn cservers configuration file, as ro o t enter:
~]# service vn cserver st art

311

Deployment G uide

You can enable the vn cserver service automatically at system start. Every time you log in,
vn cserver is automatically started. As ro o t , run
~]# ch kco n f ig vn cserver o n

14 .1.4 . T erminat ing a VNC Session
Similarly to enabling the vn cserver service, you can disable the automatic start of the service at
system start:
~]# ch kco n f ig vn cserver o f f
Or, when your system is running, you can stop the service by issuing the following command as
ro o t :
~]# service vn cserver st o p
To terminate a specific display, terminate vn cserver using the - kill option along with the display
number.

Examp le 14 .4 . T ermin at in g a Sp ecif ic D isp lay
For example, to terminate display number 2, run:
~]# vn cserver - kill :2

Examp le 14 .5. T ermin at in g an Xvn c p ro cess
If it is not possible to terminate the VNC service or display, terminate the Xvnc session using the
process ID (PID ). To view the processes, enter:
~]$ service vn cserver st at u s
Xvnc (pid 4290 4189) is running...
To terminate process 4 29 0, enter as ro o t :
~]# kill - s 15 4 29 0

14 .2. Sharing an Exist ing Deskt op
By default a logged in user has a desktop provided by X Server on display 0. A user can share their
desktop using the T ig erVN C server x0vn cserver.
Pro ced u re 14 .5. Sh arin g an X D eskt o p
To share the desktop of a logged in user, using the x0vn cserver, proceed as follows:
1. Enter the following command as ro o t

312

⁠Chapt er 1 4 . T igerVNC

~]# yu m in st all t ig ervn c- server
2. Set the VNC password for the user:
~]$ vn cp asswd
Password:
Verify:
3. Enter the following command as that user:
~]$ x0vn cserver - Passwo rd File= .vn c/p asswd - AlwaysSh ared = 1
Provided the firewall is configured to allow connections to port 59 00, the remote viewer can now
connect to display 0, and view the logged in users desktop. See Section 14.3.2.1, “ Configuring the
Firewall for VNC” for information on how to configure the firewall.

14 .3. Using a VNC Viewer
A VNC viewer is a program which shows the graphical user interface created by the VNC server and
can control the VNC server remotely. The desktop that is shared is not by default the same as the
desktop that is displayed to a user directly logged into the system. The VNC server creates a unique
desktop for every display number. Any number of clients can connect to a VNC server.

14 .3.1. Inst alling t he VNC Viewer
To install the T ig erVN C client, vn cviewer, as ro o t , run the following command:
~ ] # yu m in st all t ig ervn c
The T ig erVN C client has a graphical user interface (GUI) which can be started by entering the
command vn cviewer. Alternatively, you can operate vn cviewer through the command-line interface
(CLI). To view a list of parameters for vn cviewer enter vn cviewer - h on the command line.

14 .3.2. Connect ing t o a VNC Server
Once the VNC server is configured, you can connect to it from any VNC viewer.
Pro ced u re 14 .6 . C o n n ect in g t o a VN C Server U sin g a G U I
1. Enter the vn cviewer command with no arguments, the VN C Viewer: C o n n ect io n D et ails
utility appears. It prompts for a VNC server to connect to.
2. If required, to prevent disconnecting any existing VNC connections to the same display, select
the option to allow sharing of the desktop as follows:
a. Select the O p t io n s button.
b. Select the Misc. tab.
c. Select the Sh ared button.
d. Press O K to return to the main menu.
3. Enter an address and display number to connect to:

313

Deployment G uide

address:display_number
4. Press C o n n ect to connect to the VNC server display.
5. You will be prompted to enter the VNC password. This will be the VNC password for the user
corresponding to the display number unless a global default VNC password was set.
A window appears showing the VNC server desktop. Note that this is not the desktop the
normal user sees, it is an Xvnc desktop.
Pro ced u re 14 .7. C o n n ect in g t o a VN C Server U sin g t h e C LI
1. Enter the viewer command with the address and display number as arguments:
vncviewer address:display_number
Where address is an IP address or host name.
2. Authenticate yourself by entering the VNC password. This will be the VNC password for the
user corresponding to the display number unless a global default VNC password was set.
3. A window appears showing the VNC server desktop. Note that this is not the desktop the
normal user sees, it is the Xvnc desktop.

1 4 .3.2 .1 . Co nfiguring t he Fire wall fo r VNC
When using a non-encrypted connection, the firewall might block your connection. The VNC protocol
is remote framebuffer (RFB), which is transported in T C P packets. If required, open a port for the T C P
protocol as described below. When using the - via option, traffic is redirected over SSH which is
enabled by default.

Note
The default port of VNC server is 5900. To reach the port through which a remote desktop will
be accessible, sum the default port and the user's assigned display number. For example, for
the second display: 2 + 5900 = 5902.

Pro ced u re 14 .8. O p en in g a Po rt U sin g lo kkit
The lo kkit command provides a way to quickly enable a port using the command line.
1. To enable a specific port, for example port 59 02 for T C P, issue the following command as
ro o t :
~]# lo kkit - - p o rt = 59 02:t cp - - u p d at e
Note that this will restart the firewall as long as it has not been disabled with the - - d isab led
option. Active connections will be terminated and time out on the initiating machine.
2. Verify whether the chosen port is open. As ro o t , enter:
~]# ip t ab les - L - n | g rep ' t cp .*59 '
ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0

314

state NEW tcp dpt:5902

⁠Chapt er 1 4 . T igerVNC

3. If you are unsure of the port numbers in use for VNC, as ro o t , enter:
~]# n et st at - t n lp
tcp 0 0 0.0.0.0:6003
tcp 0 0 0.0.0.0:5900
tcp 0 0 0.0.0.0:5902
tcp 0 0 0.0.0.0:5903
tcp 0 0 0.0.0.0:6002

0.0.0.0:*
0.0.0.0:*
0.0.0.0:*
0.0.0.0:*
0.0.0.0:*

LISTEN
LISTEN
LISTEN
LISTEN
LISTEN

4290/Xvnc
7013/x0vncserver
4189/Xvnc
4290/Xvnc
4189/Xvnc

Ports starting 59 XX are for the VNC R FB protocol. Ports starting 6 0XX are for the X windows
protocol.
To list the ports and the Xvnc session's associated user, as ro o t , enter:
~]# lso f - i - P | g rep vn c
Xvnc
4189 jane 0u IPv6 27972 0t0 TCP *:6002 (LISTEN)
Xvnc
4189 jane 1u IPv4 27973 0t0 TCP *:6002 (LISTEN)
Xvnc
4189 jane 6u IPv4 27979 0t0 TCP *:5902 (LISTEN)
Xvnc
4290 joe 0u IPv6 28231 0t0 TCP *:6003 (LISTEN)
Xvnc
4290 joe 1u IPv4 28232 0t0 TCP *:6003 (LISTEN)
Xvnc
4290 joe 6u IPv4 28244 0t0 TCP *:5903 (LISTEN)
x0vncserv 7013 joe 4u IPv4 47578 0t0 TCP *:5900 (LISTEN)
Pro ced u re 14 .9 . C o n f ig u rin g t h e Firewall U sin g an Ed it o r
When preparing a configuration file for multiple installations using administration tools, it is useful to
edit the firewall configuration file directly. Note that any mistakes in the configuration file could have
unexpected consequences, cause an error, and prevent the firewall settings from being applied.
Therefore, check the /et c/sysco n f ig /syst em- co n f ig - f irewall file thoroughly after editing.
1. To check what the firewall is configured to allow, issue the following command as ro o t to
view the firewall configuration file:
~]# less /et c/sysco n f ig /syst em- co n f ig - f irewall
# Configuration file for system-config-firewall
--enabled
--service=ssh
In this example taken from a default installation, the firewall is enabled but VNC ports have
not been configured to pass through.
2. Open /et c/sysco n f ig /syst em- co n f ig - f irewall for editing as ro o t and add lines in the
following format to the firewall configuration file:
--port=port_number:tcp
For example, to add port 59 02 :
~]# vi /et c/sysco n f ig /syst em- co n f ig - f irewall
# Configuration file for system-config-firewall
--enabled
--service=ssh
- - p o rt = 59 02:t cp

315

Deployment G uide

3. Note that these changes will not take effect even if the firewall is reloaded or the system
rebooted. To apply the settings in /et c/sysco n f ig /syst em- co n f ig - f irewall, issue the
following command as ro o t :
~]# lo kkit - - u p d at e

14 .3.3. Connect ing t o VNC Server Using SSH
VNC is a clear text network protocol with no security against possible attacks on the communication.
To make the communication secure, you can encrypt your server-client connection by using the - via
option. This will create an SSH tunnel between the VNC server and the client.
The format of the command to encrypt a VNC server-client connection is as follows:
vncviewer -via user@host:display_number

Examp le 14 .6 . U sin g t h e - via O p t io n
1. To connect to a VNC server using SSH , enter a command as follows:
$ vn cviewer - via jo e@ 19 2.16 8.2.101 127.0.0.1:3
2. When you are prompted to, type the password, and confirm by pressing En t er.
3. A window with a remote desktop appears on your screen.
For more information on using SSH , see Chapter 13, OpenSSH.

14 .4 . Addit ional Resources
For more information about TigerVNC, see the resources listed below.

Inst alled Document at ion
vn cserver( 1) — The manual page for the VNC server utility.
vn cviewer( 1) — The manual page for the VNC viewer.
vn cp asswd ( 1) — The manual page for the VNC password command.
Xvn c( 1) — The manual page for the Xvnc server configuration options.
x0vn cserver( 1) — The manual page for the T ig erVN C server for sharing existing X servers.

316

⁠P art VI. Servers

⁠Part VI. Servers
This part discusses various topics related to servers such as how to set up a Web server or share
files and directories over the network.

317

Deployment G uide

Chapter 15. DHCP Servers
D ynamic Host Configuration Protocol (D HCP) is a network protocol that automatically assigns
TCP/IP information to client machines. Each D HCP client connects to the centrally located D HCP
server, which returns the network configuration (including the IP address, gateway, and D NS
servers) of that client.

15.1. Why Use DHCP?
D HCP is useful for automatic configuration of client network interfaces. When configuring the client
system, you can choose D HCP instead of specifying an IP address, netmask, gateway, or D NS
servers. The client retrieves this information from the D HCP server. D HCP is also useful if you want to
change the IP addresses of a large number of systems. Instead of reconfiguring all the systems, you
can just edit one configuration file on the server for the new set of IP addresses. If the D NS servers for
an organization changes, the changes happen on the D HCP server, not on the D HCP clients. When
you restart the network or reboot the clients, the changes go into effect.
If an organization has a functional D HCP server correctly connected to a network, laptops and other
mobile computer users can move these devices from office to office.

15.2. Configuring a DHCP Server
The dhcp package contains an Internet Systems Consortium (ISC) D HCP server. First, install the
package as the superuser:
~]# yu m in st all d h cp
Installing the dhcp package creates a file, /et c/d h cp /d h cp d .co n f , which is merely an empty
configuration file:
~]# cat /et c/d h cp /d h cp d .co n f
#
# D HCP Server Configuration file.
# see /usr/share/doc/dhcp*/dhcpd.conf.sample
The sample configuration file can be found at /u sr/sh are/d o c/d h cp /d h cp d .co n f .samp le. You should use this file to help you configure
/et c/d h cp /d h cp d .co n f , which is explained in detail below.
D HCP also uses the file /var/lib /d h cp d /d h cp d .leases to store the client lease database. See
Section 15.2.2, “ Lease D atabase” for more information.

15.2.1. Configurat ion File
The first step in configuring a D HCP server is to create the configuration file that stores the network
information for the clients. Use this file to declare options and global options for client systems.
The configuration file can contain extra tabs or blank lines for easier formatting. Keywords are caseinsensitive and lines beginning with a hash sign (#) are considered comments.
There are two types of statements in the configuration file:

318

⁠Chapt er 1 5. DHCP Servers

Parameters — State how to perform a task, whether to perform a task, or what network
configuration options to send to the client.
D eclarations — D escribe the topology of the network, describe the clients, provide addresses for
the clients, or apply a group of parameters to a group of declarations.
The parameters that start with the keyword option are referred to as options. These options control
D HCP options; whereas, parameters configure values that are not optional or control how the D HCP
server behaves.
Parameters (including options) declared before a section enclosed in curly brackets ({ }) are
considered global parameters. Global parameters apply to all the sections below it.

Restart the DHCP Daemon for the Changes to T ake Effect
If the configuration file is changed, the changes do not take effect until the D HCP daemon is
restarted with the command service d h cp d rest art .

Use the omshell Command
Instead of changing a D HCP configuration file and restarting the service each time, using the
o msh ell command provides an interactive way to connect to, query, and change the
configuration of a D HCP server. By using o msh ell, all changes can be made while the server
is running. For more information on o msh ell, see the o msh ell man page.
In Example 15.1, “ Subnet D eclaration” , the ro u t ers, su b n et - mask, d o main - search , d o main n ame- servers, and t ime- o f f set options are used for any h o st statements declared below it.
For every su b n et which will be served, and for every su b n et to which the D HCP server is
connected, there must be one su b n et declaration, which tells the D HCP daemon how to recognize
that an address is on that su b n et . A su b n et declaration is required for each su b n et even if no
addresses will be dynamically allocated to that su b n et .
In this example, there are global options for every D HCP client in the subnet and a ran g e declared.
Clients are assigned an IP address within the ran g e.

Examp le 15.1. Su b n et D eclarat io n
subnet 192.168.1.0 netmask 255.255.255.0 {
option routers
192.168.1.254;
option subnet-mask
255.255.255.0;
option domain-search
" example.com" ;
option domain-name-servers
192.168.1.1;
option time-offset
-18000; # Eastern Standard Time
range 192.168.1.10 192.168.1.100;
}

To configure a D HCP server that leases a dynamic IP address to a system within a subnet, modify
Example 15.2, “ Range Parameter” with your values. It declares a default lease time, maximum lease
time, and network configuration values for the clients. This example assigns IP addresses in the
ran g e 192.168.1.10 and 192.168.1.100 to client systems.

319

Deployment G uide

Examp le 15.2. R an g e Paramet er
default-lease-time 600;
max-lease-time 7200;
option subnet-mask 255.255.255.0;
option broadcast-address 192.168.1.255;
option routers 192.168.1.254;
option domain-name-servers 192.168.1.1, 192.168.1.2;
option domain-search " example.com" ;
subnet 192.168.1.0 netmask 255.255.255.0 {
range 192.168.1.10 192.168.1.100;
}

To assign an IP address to a client based on the MAC address of the network interface card, use the
h ard ware et h ern et parameter within a h o st declaration. As demonstrated in Example 15.3, “ Static
IP Address Using D HCP” , the h o st ap ex declaration specifies that the network interface card with
the MAC address 00:A0:78:8E:9E:AA always receives the IP address 192.168.1.4.
Note that you can also use the optional parameter h o st - n ame to assign a host name to the client.

Examp le 15.3. St at ic IP Ad d ress U sin g D H C P
host apex {
option host-name " apex.example.com" ;
hardware ethernet 00:A0:78:8E:9E:AA;
fixed-address 192.168.1.4;
}

All subnets that share the same physical network should be declared within a sh ared - n et wo rk
declaration as shown in Example 15.4, “ Shared-network D eclaration” . Parameters within the
sh ared - n et wo rk, but outside the enclosed su b n et declarations, are considered to be global
parameters. The name of the sh ared - n et wo rk must be a descriptive title for the network, such as
using the title 'test-lab' to describe all the subnets in a test lab environment.

Examp le 15.4 . Sh ared - n et wo rk D eclarat io n
shared-network name {
option domain-search
" test.redhat.com" ;
option domain-name-servers
ns1.redhat.com, ns2.redhat.com;
option routers
192.168.0.254;
#more parameters for EXAMPLE shared-network
subnet 192.168.1.0 netmask 255.255.252.0 {
#parameters for subnet
range 192.168.1.1 192.168.1.254;
}
subnet 192.168.2.0 netmask 255.255.252.0 {
#parameters for subnet
range 192.168.2.1 192.168.2.254;
}
}

320

⁠Chapt er 1 5. DHCP Servers

As demonstrated in Example 15.5, “ Group D eclaration” , the g ro u p declaration is used to apply
global parameters to a group of declarations. For example, shared networks, subnets, and hosts can
be grouped.

Examp le 15.5. G ro u p D eclarat io n
group {
option routers
192.168.1.254;
option subnet-mask
255.255.255.0;
option domain-search
" example.com" ;
option domain-name-servers
192.168.1.1;
option time-offset
-18000; # Eastern Standard Time
host apex {
option host-name " apex.example.com" ;
hardware ethernet 00:A0:78:8E:9E:AA;
fixed-address 192.168.1.4;
}
host raleigh {
option host-name " raleigh.example.com" ;
hardware ethernet 00:A1:D D :74:C3:F2;
fixed-address 192.168.1.6;
}
}

Using the Sample Configuration File
You can use the provided sample configuration file as a starting point and add custom
configuration options to it. To copy this file to the proper location, use the following command
as ro o t :
~]# cp /u sr/sh are/d o c/d h cp - /d h cp d .co n f .samp le
/et c/d h cp /d h cp d .co n f
... where  is the D HCP version number.

For a complete list of option statements and what they do, see the d h cp - o p t io n s man page.

15.2.2. Lease Dat abase
On the D HCP server, the file /var/lib /d h cp d /d h cp d .leases stores the D HCP client lease database.
D o not change this file. D HCP lease information for each recently assigned IP address is
automatically stored in the lease database. The information includes the length of the lease, to whom
the IP address has been assigned, the start and end dates for the lease, and the MAC address of the
network interface card that was used to retrieve the lease.
All times in the lease database are in Coordinated Universal Time (UTC), not local time.

321

Deployment G uide

The lease database is recreated from time to time so that it is not too large. First, all known leases are
saved in a temporary lease database. The d h cp d .leases file is renamed d h cp d .leases~ and the
temporary lease database is written to d h cp d .leases.
The D HCP daemon could be killed or the system could crash after the lease database has been
renamed to the backup file but before the new file has been written. If this happens, the
d h cp d .leases file does not exist, but it is required to start the service. D o not create a new lease file.
If you do, all old leases are lost which causes many problems. The correct solution is to rename the
d h cp d .leases~ backup file to d h cp d .leases and then start the daemon.

15.2.3. St art ing and St opping t he Server

Starting the DHCP Server for the First T ime
When the D HCP server is started for the first time, it fails unless the d h cp d .leases file exists.
Use the command t o u ch /var/lib /d h cp d /d h cp d .leases to create the file if it does not exist.
If the same server is also running BIND as a D NS server, this step is not necessary, as starting
the n amed service automatically checks for a d h cp d .leases file.

To start the D HCP service, use the command /sb in /service d h cp d st art . To stop the D HCP server,
use the command /sb in /service d h cp d st o p .
By default, the D HCP service does not start at boot time. For information on how to configure the
daemon to start automatically at boot time, see Chapter 11, Services and Daemons.
If more than one network interface is attached to the system, but the D HCP server should only be
started on one of the interfaces, configure the D HCP server to start only on that device. In
/et c/sysco n f ig /d h cp d , add the name of the interface to the list of D H C PD AR G S:
# Command line options here
D HCPD ARGS=eth0
This is useful for a firewall machine with two network cards. One network card can be configured as a
D HCP client to retrieve an IP address to the Internet. The other network card can be used as a D HCP
server for the internal network behind the firewall. Specifying only the network card connected to the
internal network makes the system more secure because users can not connect to the daemon via the
Internet.
Other command-line options that can be specified in /et c/sysco n f ig /d h cp d include:
- p  — Specifies the UD P port number on which d h cp d should listen. The default is
port 67. The D HCP server transmits responses to the D HCP clients at a port number one greater
than the UD P port specified. For example, if the default port 67 is used, the server listens on port
67 for requests and responds to the client on port 68. If a port is specified here and the D HCP
relay agent is used, the same port on which the D HCP relay agent should listen must be specified.
See Section 15.2.4, “ D HCP Relay Agent” for details.
- f — Runs the daemon as a foreground process. This is mostly used for debugging.
- d — Logs the D HCP server daemon to the standard error descriptor. This is mostly used for
debugging. If this is not specified, the log is written to /var/lo g /messag es.

322

⁠Chapt er 1 5. DHCP Servers

- cf  — Specifies the location of the configuration file. The default location is
/et c/d h cp /d h cp d .co n f .
- lf  — Specifies the location of the lease database file. If a lease database file already
exists, it is very important that the same file be used every time the D HCP server is started. It is
strongly recommended that this option only be used for debugging purposes on non-production
machines. The default location is /var/lib /d h cp d /d h cp d .leases.
- q — D o not print the entire copyright message when starting the daemon.

15.2.4 . DHCP Relay Agent
The D HCP Relay Agent (d h crelay) allows for the relay of D HCP and BOOTP requests from a subnet
with no D HCP server on it to one or more D HCP servers on other subnets.
When a D HCP client requests information, the D HCP Relay Agent forwards the request to the list of
D HCP servers specified when the D HCP Relay Agent is started. When a D HCP server returns a reply,
the reply is broadcast or unicast on the network that sent the original request.
The D HCP Relay Agent listens for D HCP requests on all interfaces unless the interfaces are specified
in /et c/sysco n f ig /d h crelay with the IN T ER FAC ES directive.
To start the D HCP Relay Agent, use the command service d h crelay st art .

15.3. Configuring a DHCP Client
To configure a D HCP client manually, modify the /et c/sysco n f ig /n et wo rk file to enable networking
and the configuration file for each network device in the /et c/sysco n f ig /n et wo rk- scrip t s directory.
In this directory, each device should have a configuration file named if cf g - et h 0, where et h 0 is the
network device name.
Make sure that the /et c/sysco n f ig /n et wo rk- scrip t s/if cf g - et h 0 file contains the following lines:
D EVICE=eth0
BOOTPROTO=dhcp
ONBOOT=yes
To use D HCP, set a configuration file for each device.
Other options for the network script include:
D H C P_H O ST N AME — Only use this option if the D HCP server requires the client to specify a
host name before receiving an IP address.
PEER D N S= , where  is one of the following:
yes — Modify /et c/reso lv.co n f with information from the server. This is the default.
n o — D o not modify /et c/reso lv.co n f .
If you prefer using a graphical interface, see Chapter 9, NetworkManager for instructions on using
N et wo rkMan ag er to configure a network interface to use D HCP.

323

Deployment G uide

Advanced configurations
For advanced configurations of client D HCP options such as protocol timing, lease
requirements and requests, dynamic D NS support, aliases, as well as a wide variety of values
to override, prepend, or append to client-side configurations, see the d h clien t and
d h clien t .co n f man pages.

15.4 . Configuring a Mult ihomed DHCP Server
A multihomed D HCP server serves multiple networks, that is, multiple subnets. The examples in these
sections detail how to configure a D HCP server to serve multiple networks, select which network
interfaces to listen on, and how to define network settings for systems that move networks.
Before making any changes, back up the existing /et c/sysco n f ig /d h cp d and
/et c/d h cp /d h cp d .co n f files.
The D HCP daemon listens on all network interfaces unless otherwise specified. Use the
/et c/sysco n f ig /d h cp d file to specify which network interfaces the D HCP daemon listens on. The
following /et c/sysco n f ig /d h cp d example specifies that the D HCP daemon listens on the et h 0 and
et h 1 interfaces:
D HCPD ARGS=" eth0 eth1" ;
If a system has three network interfaces cards — et h 0, et h 1, and et h 2 — and it is only desired that
the D HCP daemon listens on the et h 0 card, then only specify et h 0 in /et c/sysco n f ig /d h cp d :
D HCPD ARGS=" eth0" ;
The following is a basic /et c/d h cp /d h cp d .co n f file, for a server that has two network interfaces,
et h 0 in a 10.0.0.0/24 network, and et h 1 in a 172.16.0.0/24 network. Multiple su b n et declarations
allow you to define different settings for multiple networks:
default-lease-time 600;
max-lease-time 7200;
subnet 10.0.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 10.0.0.1;
range 10.0.0.5 10.0.0.15;
}
subnet 172.16.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 172.16.0.1;
range 172.16.0.5 172.16.0.15;
}
su b n et 10.0.0.0 n et mask 255.255.255.0;
A su b n et declaration is required for every network your D HCP server is serving. Multiple
subnets require multiple su b n et declarations. If the D HCP server does not have a network
interface in a range of a su b n et declaration, the D HCP server does not serve that network.

324

⁠Chapt er 1 5. DHCP Servers

If there is only one su b n et declaration, and no network interfaces are in the range of that
subnet, the D HCP daemon fails to start, and an error such as the following is logged to
/var/lo g /messag es:
dhcpd: No subnet declaration for eth0 (0.0.0.0).
dhcpd: ** Ignoring requests on eth0. If this is not what
dhcpd: you want, please write a subnet declaration
dhcpd: in your dhcpd.conf file for the network segment
dhcpd: to which interface eth1 is attached. **
dhcpd:
dhcpd:
dhcpd: Not configured to listen on any interfaces!
o p t io n su b n et - mask 255.255.255.0;
The o p t io n su b n et - mask option defines a subnet mask, and overrides the n et mask
value in the su b n et declaration. In simple cases, the subnet and netmask values are the
same.
o p t io n ro u t ers 10.0.0.1;
The o p t io n ro u t ers option defines the default gateway for the subnet. This is required for
systems to reach internal networks on a different subnet, as well as external networks.
ran g e 10.0.0.5 10.0.0.15;
The ran g e option specifies the pool of available IP addresses. Systems are assigned an
address from the range of specified IP addresses.
For further information, see the d h cp d .co n f ( 5) man page.

15.4 .1. Host Configurat ion
Before making any changes, back up the existing /et c/sysco n f ig /d h cp d and
/et c/d h cp /d h cp d .co n f files.
C o n f ig u rin g a Sin g le Syst em f o r Mu lt ip le N et wo rks
The following /et c/d h cp /d h cp d .co n f example creates two subnets, and configures an IP address
for the same system, depending on which network it connects to:
default-lease-time 600;
max-lease-time 7200;
subnet 10.0.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 10.0.0.1;
range 10.0.0.5 10.0.0.15;
}
subnet 172.16.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 172.16.0.1;
range 172.16.0.5 172.16.0.15;
}
host example0 {
hardware ethernet 00:1A:6B:6A:2E:0B;
fixed-address 10.0.0.20;
}

325

Deployment G uide

host example1 {
hardware ethernet 00:1A:6B:6A:2E:0B;
fixed-address 172.16.0.20;
}
h o st example0
The h o st declaration defines specific parameters for a single system, such as an IP
address. To configure specific parameters for multiple hosts, use multiple h o st
declarations.
Most D HCP clients ignore the name in h o st declarations, and as such, this name can be
anything, as long as it is unique to other h o st declarations. To configure the same system
for multiple networks, use a different name for each h o st declaration, otherwise the D HCP
daemon fails to start. Systems are identified by the h ard ware et h ern et option, not the
name in the h o st declaration.
h ard ware et h ern et 00:1A:6B:6A:2E:0B;
The h ard ware et h ern et option identifies the system. To find this address, run the ip lin k
command.
f ixed - ad d ress 10.0.0.20;
The f ixed - ad d ress option assigns a valid IP address to the system specified by the
h ard ware et h ern et option. This address must be outside the IP address pool specified
with the ran g e option.
If o p t io n statements do not end with a semicolon, the D HCP daemon fails to start, and an error such
as the following is logged to /var/lo g /messag es:
/etc/dhcp/dhcpd.conf line 20: semicolon expected.
dhcpd: }
dhcpd: ^
dhcpd: /etc/dhcp/dhcpd.conf line 38: unexpected end of file
dhcpd:
dhcpd: ^
dhcpd: Configuration file errors encountered -- exiting
C o n f ig u rin g Syst ems wit h Mu lt ip le N et wo rk In t erf aces
The following h o st declarations configure a single system, which has multiple network interfaces, so
that each interface receives the same IP address. This configuration will not work if both network
interfaces are connected to the same network at the same time:
host interface0 {
hardware ethernet 00:1a:6b:6a:2e:0b;
fixed-address 10.0.0.18;
}
host interface1 {
hardware ethernet 00:1A:6B:6A:27:3A;
fixed-address 10.0.0.18;
}
For this example, in t erf ace0 is the first network interface, and in t erf ace1 is the second interface.
The different h ard ware et h ern et options identify each interface.

326

⁠Chapt er 1 5. DHCP Servers

If such a system connects to another network, add more h o st declarations, remembering to:
assign a valid f ixed - ad d ress for the network the host is connecting to.
make the name in the h o st declaration unique.
When a name given in a h o st declaration is not unique, the D HCP daemon fails to start, and an
error such as the following is logged to /var/lo g /messag es:
dhcpd: /etc/dhcp/dhcpd.conf line 31: host interface0: already exists
dhcpd: }
dhcpd: ^
dhcpd: Configuration file errors encountered -- exiting
This error was caused by having multiple h o st in t erf ace0 declarations defined in
/et c/d h cp /d h cp d .co n f .

15.5. DHCP for IPv6 (DHCPv6)
The ISC D HCP includes support for IPv6 (D HCPv6) since the 4.x release with a D HCPv6 server,
client and relay agent functionality. The server, client and relay agents support both IPv4 and IPv6.
However, the client and the server can only manage one protocol at a time — for dual support they
must be started separately for IPv4 and IPv6.
The D HCPv6 server configuration file can be found at /et c/d h cp /d h cp d 6 .co n f .
The sample server configuration file can be found at /u sr/sh are/d o c/d h cp /d h cp d 6 .co n f .samp le.
To start the D HCPv6 service, use the command /sb in /service d h cp d 6 st art .
A simple D HCPv6 server configuration file can look like this:
subnet6 2001:db8:0:1::/64 {
range6 2001:db8:0:1::129 2001:db8:0:1::254;
option dhcp6.name-servers fec0:0:0:1::1;
option dhcp6.domain-search " domain.example" ;
}

15.6. Addit ional Resources
For additional information, see The DHCP Handbook; Ralph Droms and Ted Lemon; 2003 or the
following resources.

15.6.1. Inst alled Document at ion
d h cp d man page — D escribes how the D HCP daemon works.
d h cp d .co n f man page — Explains how to configure the D HCP configuration file; includes some
examples.
d h cp d .leases man page — D escribes a persistent database of leases.
d h cp - o p t io n s man page — Explains the syntax for declaring D HCP options in d h cp d .co n f ;
includes some examples.

327

Deployment G uide

d h crelay man page — Explains the D HCP Relay Agent and its configuration options.
/u sr/sh are/d o c/d h cp - / — Contains sample files, READ ME files, and release notes for
current versions of the D HCP service.

328

⁠Chapt er 1 6 . DNS Servers

Chapter 16. DNS Servers
D N S (D omain Name System), also known as a nameserver, is a network system that associates host
names with their respective IP addresses. For users, this has the advantage that they can refer to
machines on the network by names that are usually easier to remember than the numerical network
addresses. For system administrators, using the nameserver allows them to change the IP address
for a host without ever affecting the name-based queries, or to decide which machines handle these
queries.

16.1. Int roduct ion t o DNS
D NS is usually implemented using one or more centralized servers that are authoritative for certain
domains. When a client host requests information from a nameserver, it usually connects to port 53.
The nameserver then attempts to resolve the name requested. If it does not have an authoritative
answer, or does not already have the answer cached from an earlier query, it queries other
nameservers, called root nameservers, to determine which nameservers are authoritative for the name
in question, and then queries them to get the requested name.

16.1.1. Nameserver Zones
In a D NS server such as BIND (Berkeley Internet Name D omain), all information is stored in basic
data elements called resource records (RR). The resource record is usually a fully qualified domain name
(FQD N) of a host, and is broken down into multiple sections organized into a tree-like hierarchy. This
hierarchy consists of a main trunk, primary branches, secondary branches, and so on.

Examp le 16 .1. A simp le reso u rce reco rd
bob.sales.example.com

Each level of the hierarchy is divided by a period (that is, .). In Example 16.1, “ A simple resource
record” , co m defines the top-level domain, examp le its subdomain, and sales the subdomain of
examp le. In this case, b o b identifies a resource record that is part of the sales.examp le.co m
domain. With the exception of the part furthest to the left (that is, b o b ), each of these sections is
called a zone and defines a specific namespace.
Z ones are defined on authoritative nameservers through the use of zone files, which contain
definitions of the resource records in each zone. Z one files are stored on primary nameservers (also
called master nameservers), where changes are made to the files, and secondary nameservers (also
called slave nameservers), which receive zone definitions from the primary nameservers. Both primary
and secondary nameservers are authoritative for the zone and look the same to clients. D epending
on the configuration, any nameserver can also serve as a primary or secondary server for multiple
zones at the same time.

16.1.2. Nameserver T ypes
There are two nameserver configuration types:
au t h o rit at ive
Authoritative nameservers answer to resource records that are part of their zones only. This
category includes both primary (master) and secondary (slave) nameservers.

329

Deployment G uide

recu rsive
Recursive nameservers offer resolution services, but they are not authoritative for any zone.
Answers for all resolutions are cached in a memory for a fixed period of time, which is
specified by the retrieved resource record.
Although a nameserver can be both authoritative and recursive at the same time, it is recommended
not to combine the configuration types. To be able to perform their work, authoritative servers should
be available to all clients all the time. On the other hand, since the recursive lookup takes far more
time than authoritative responses, recursive servers should be available to a restricted number of
clients only, otherwise they are prone to distributed denial of service (D D oS) attacks.

16.1.3. BIND as a Nameserver
BIND consists of a set of D NS-related programs. It contains a nameserver called n amed , an
administration utility called rn d c, and a debugging tool called d ig . See Chapter 11, Services and
Daemons for more information on how to run a service in Red Hat Enterprise Linux.

16.2. BIND
This chapter covers B IN D (Berkeley Internet Name D omain), the D NS server included in Red Hat
Enterprise Linux. It focuses on the structure of its configuration files, and describes how to administer
it both locally and remotely.

16.2.1. Configuring t he named Service
When the n amed service is started, it reads the configuration from the files as described in
Table 16.1, “ The named service configuration files” .
T ab le 16 .1. T h e n amed service co n f ig u rat io n f iles
Pat h

D escrip t io n

/et c/n amed .co n f
/et c/n amed /

The main configuration file.
An auxiliary directory for configuration files that are included
in the main configuration file.

The configuration file consists of a collection of statements with nested options surrounded by
opening and closing curly brackets. Note that when editing the file, you have to be careful not to
make any syntax error, otherwise the n amed service will not start. A typical /et c/n amed .co n f file is
organized as follows:
statement-1 [" statement-1-name" ] [statement-1-class] {
option-1;
option-2;
option-N;
};
statement-2 [" statement-2-name" ] [statement-2-class] {
option-1;
option-2;
option-N;
};
statement-N [" statement-N-name" ] [statement-N-class] {

330

⁠Chapt er 1 6 . DNS Servers

option-1;
option-2;
option-N;
};

Running BIND in a chroot environment
If you have installed the bind-chroot package, the BIND service will run in the
/var/n amed /ch ro o t environment. In that case, the initialization script will mount the above
configuration files using the mo u n t - - b in d command, so that you can manage the
configuration outside this environment. There is no need to copy anything into the
/var/n amed /ch ro o t directory because it is mounted automatically. This simplifies
maintenance since you do not need to take any special care of B IN D configuration files if it is
run in a ch ro o t environment. You can organize everything as you would with B IN D not
running in a ch ro o t environment.
The following directories are automatically mounted into /var/n amed /ch ro o t if they are empty
in the /var/n amed /ch ro o t directory. They must be kept empty if you want them to be mounted
into /var/n amed /ch ro o t :
/var/n amed
/et c/p ki/d n ssec- keys
/et c/n amed
/u sr/lib 6 4 /b in d or /u sr/lib /b in d (architecture dependent).
The following files are also mounted if the target file does not exist in /var/n amed /ch ro o t .
/et c/n amed .co n f
/et c/rn d c.co n f
/et c/rn d c.key
/et c/n amed .rf c19 12.z o n es
/et c/n amed .d n ssec.keys
/et c/n amed .iscd lv.key
/et c/n amed .ro o t .key

1 6 .2 .1 .1 . Co m m o n St at e m e nt T ype s
The following types of statements are commonly used in /et c/n amed .co n f :
acl
The acl (Access Control List) statement allows you to define groups of hosts, so that they
can be permitted or denied access to the nameserver. It takes the following form:
acl acl-name {
match-element;
...
};
The acl-name statement name is the name of the access control list, and the match-element
option is usually an individual IP address (such as 10.0.1.1) or a CID R (Classless InterD omain Routing) network notation (for example, 10.0.1.0/24 ). For a list of already defined
keywords, see Table 16.2, “ Predefined access control lists” .

331

Deployment G uide

T ab le 16 .2. Pred ef in ed access co n t ro l list s
K eywo rd

D escrip t io n

an y
lo calh o st
lo caln et s

Matches every IP address.
Matches any IP address that is in use by the local system.
Matches any IP address on any network to which the local system
is connected.
D oes not match any IP address.

none

The acl statement can be especially useful in conjunction with other statements such as
o p t io n s. Example 16.2, “ Using acl in conjunction with options” defines two access control
lists, b lack- h at s and red - h at s, and adds b lack- h at s on the blacklist while granting
red - h at s a normal access.

Examp le 16 .2. U sin g acl in co n ju n ct io n wit h o p t io n s
acl black-hats {
10.0.2.0/24;
192.168.0.0/24;
1234:5678::9abc/24;
};
acl red-hats {
10.0.1.0/24;
};
options {
blackhole { black-hats; };
allow-query { red-hats; };
allow-query-cache { red-hats; };
};
in clu d e
The in clu d e statement allows you to include files in the /et c/n amed .co n f , so that
potentially sensitive data can be placed in a separate file with restricted permissions. It
takes the following form:
include " file-name"
The file-name statement name is an absolute path to a file.

Examp le 16 .3. In clu d in g a f ile t o /et c/n amed .co n f
include " /etc/named.rfc1912.zones" ;
o p t io n s
The o p t io n s statement allows you to define global server configuration options as well as
to set defaults for other statements. It can be used to specify the location of the n amed
working directory, the types of queries allowed, and much more. It takes the following form:

332

⁠Chapt er 1 6 . DNS Servers

options {
option;
...
};
For a list of frequently used option directives, see Table 16.3, “ Commonly used options”
below.
T ab le 16 .3. C o mmo n ly u sed o p t io n s
O p t io n

D escrip t io n

allo w- q u ery

Specifies which hosts are allowed to query the nameserver for
authoritative resource records. It accepts an access control list, a
collection of IP addresses, or networks in the CID R notation. All
hosts are allowed by default.
Specifies which hosts are allowed to query the nameserver for
non-authoritative data such as recursive queries. Only lo calh o st
and lo caln et s are allowed by default.
Specifies which hosts are not allowed to query the nameserver.
This option should be used when particular host or network floods
the server with requests. The default option is n o n e.
Specifies a working directory for the n amed service. The default
option is /var/n amed /.
Specifies whether to return D NSSEC related resource records. The
default option is yes.
Specifies whether to prove that resource records are authentic via
D NSSEC. The default option is yes.
Specifies a list of valid IP addresses for nameservers to which the
requests should be forwarded for resolution.
Specifies the behavior of the f o rward ers directive. It accepts the
following options:

allo w- q u erycach e
b lackh o le

d irect o ry
d n ssec- en ab le
d n ssecvalid at io n
f o rward ers
f o rward

f irst — The server will query the nameservers listed in the
f o rward ers directive before attempting to resolve the name on
its own.
o n ly — When unable to query the nameservers listed in the
f o rward ers directive, the server will not attempt to resolve the
name on its own.
list en - o n

list en - o n - v6

max- cach e- siz e

Specifies the IPv4 network interface on which to listen for queries.
On a D NS server that also acts as a gateway, you can use this
option to answer queries originating from a single network only.
All IPv4 interfaces are used by default.
Specifies the IPv6 network interface on which to listen for queries.
On a D NS server that also acts as a gateway, you can use this
option to answer queries originating from a single network only.
All IPv6 interfaces are used by default.
Specifies the maximum amount of memory to be used for server
caches. When the limit is reached, the server causes records to
expire prematurely so that the limit is not exceeded. In a server with
multiple views, the limit applies separately to the cache of each
view. The default option is 32M.

333

Deployment G uide

O p t io n

D escrip t io n

n o t if y

Specifies whether to notify the secondary nameservers when a
zone is updated. It accepts the following options:
yes — The server will notify all secondary nameservers.
n o — The server will not notify any secondary nameserver.
mast er- o n ly — The server will notify primary server for the
zone only.
exp licit — The server will notify only the secondary servers
that are specified in the also - n o t if y list within a zone
statement.

p id - f ile
recu rsio n
st at ist ics- f ile

Specifies the location of the process ID file created by the n amed
service.
Specifies whether to act as a recursive server. The default option is
yes.
Specifies an alternate location for statistics files. The
/var/n amed /n amed .st at s file is used by default.

Restrict recursive servers to selected clients only
To prevent distributed denial of service (D D oS) attacks, it is recommended that you
use the allo w- q u ery- cach e option to restrict recursive D NS services for a
particular subset of clients only.
See the BIND 9 Administrator Reference Manual referenced in Section 16.2.7.1, “ Installed
D ocumentation” , and the n amed .co n f manual page for a complete list of available
options.

Examp le 16 .4 . U sin g t h e o p t io n s st at emen t
options {
allow-query
{ localhost; };
listen-on port 53 { 127.0.0.1; };
listen-on-v6 port 53 { ::1; };
max-cache-size 256M;
directory
" /var/named" ;
statistics-file " /var/named/data/named_stats.txt" ;
recursion
yes;
dnssec-enable yes;
dnssec-validation yes;
};
zone
The z o n e statement allows you to define the characteristics of a zone, such as the location
of its configuration file and zone-specific options, and can be used to override the global
o p t io n s statements. It takes the following form:

334

⁠Chapt er 1 6 . DNS Servers

zone zone-name [zone-class] {
option;
...
};
The zone-name attribute is the name of the zone, zone-class is the optional class of the zone,
and option is a z o n e statement option as described in Table 16.4, “ Commonly used
options” .
The zone-name attribute is particularly important, as it is the default value assigned for the
$O R IG IN directive used within the corresponding zone file located in the /var/n amed /
directory. The n amed daemon appends the name of the zone to any non-fully qualified
domain name listed in the zone file. For example, if a z o n e statement defines the
namespace for examp le.co m, use examp le.co m as the zone-name so that it is placed at
the end of host names within the examp le.co m zone file.
For more information about zone files, see Section 16.2.2, “ Editing Z one Files” .
T ab le 16 .4 . C o mmo n ly u sed o p t io n s
O p t io n

D escrip t io n

allo w- q u ery

Specifies which clients are allowed to request information about
this zone. This option overrides global allo w- q u ery option. All
query requests are allowed by default.
Specifies which secondary servers are allowed to request a
transfer of the zone's information. All transfer requests are allowed
by default.
Specifies which hosts are allowed to dynamically update
information in their zone. The default option is to deny all dynamic
update requests.

allo w- t ran sf er

allo w- u p d at e

Note that you should be careful when allowing hosts to update
information about their zone. D o not set IP addresses in this
option unless the server is in the trusted network. Instead, use
TSIG key as described in Section 16.2.5.3, “ Transaction
SIGnatures (TSIG)” .
f ile
mast ers

n o t if y

Specifies the name of the file in the n amed working directory that
contains the zone's configuration data.
Specifies from which IP addresses to request authoritative zone
information. This option is used only if the zone is defined as t yp e
slave.
Specifies whether to notify the secondary nameservers when a
zone is updated. It accepts the following options:
yes — The server will notify all secondary nameservers.
n o — The server will not notify any secondary nameserver.
mast er- o n ly — The server will notify primary server for the
zone only.
exp licit — The server will notify only the secondary servers
that are specified in the also - n o t if y list within a zone
statement.

335

Deployment G uide

O p t io n

D escrip t io n

t yp e

Specifies the zone type. It accepts the following options:
d eleg at io n - o n ly — Enforces the delegation status of
infrastructure zones such as COM, NET, or ORG. Any answer
that is received without an explicit or implicit delegation is
treated as N XD O MAIN . This option is only applicable in TLD s
(Top-Level D omain) or root zone files used in recursive or
caching implementations.
f o rward — Forwards all requests for information about this
zone to other nameservers.
h in t — A special type of zone used to point to the root
nameservers which resolve queries when a zone is not
otherwise known. No configuration beyond the default is
necessary with a h in t zone.
mast er — D esignates the nameserver as authoritative for this
zone. A zone should be set as the mast er if the zone's
configuration files reside on the system.
slave — D esignates the nameserver as a slave server for this
zone. Master server is specified in mast ers directive.

Most changes to the /et c/n amed .co n f file of a primary or secondary nameserver involve
adding, modifying, or deleting z o n e statements, and only a small subset of z o n e statement
options is usually needed for a nameserver to work efficiently.
In Example 16.5, “ A zone statement for a primary nameserver” , the zone is identified as
examp le.co m, the type is set to mast er, and the n amed service is instructed to read the
/var/n amed /examp le.co m.z o n e file. It also allows only a secondary nameserver
(19 2.16 8.0.2) to transfer the zone.

Examp le 16 .5. A z o n e st at emen t f o r a p rimary n ameserver
zone " example.com" IN {
type master;
file " example.com.zone" ;
allow-transfer { 192.168.0.2; };
};

A secondary server's z o n e statement is slightly different. The type is set to slave, and the
mast ers directive is telling n amed the IP address of the master server.
In Example 16.6, “ A zone statement for a secondary nameserver” , the n amed service is
configured to query the primary server at the 19 2.16 8.0.1 IP address for information about
the examp le.co m zone. The received information is then saved to the
/var/n amed /slaves/examp le.co m.z o n e file. Note that you have to put all slave zones to
/var/n amed /slaves directory, otherwise the service will fail to transfer the zone.

Examp le 16 .6 . A z o n e st at emen t f o r a seco n d ary n ameserver
zone " example.com" {

336

⁠Chapt er 1 6 . DNS Servers

type slave;
file " slaves/example.com.zone" ;
masters { 192.168.0.1; };
};

1 6 .2 .1 .2 . Ot he r St at e m e nt T ype s
The following types of statements are less commonly used in /et c/n amed .co n f :
co n t ro ls
The co n t ro ls statement allows you to configure various security requirements necessary
to use the rn d c command to administer the n amed service.
See Section 16.2.3, “ Using the rndc Utility” for more information on the rn d c utility and its
usage.
key
The key statement allows you to define a particular key by name. Keys are used to
authenticate various actions, such as secure updates or the use of the rn d c command.
Two options are used with key:
alg o rit h m algorithm-name — The type of algorithm to be used (for example, h macmd 5).
secret "key-value" — The encrypted key.
See Section 16.2.3, “ Using the rndc Utility” for more information on the rn d c utility and its
usage.
lo g g in g
The lo g g in g statement allows you to use multiple types of logs, so called channels. By
using the ch an n el option within the statement, you can construct a customized type of log
with its own file name (f ile), size limit (siz e), versioning (versio n ), and level of importance
(severit y). Once a customized channel is defined, a cat eg o ry option is used to categorize
the channel and begin logging when the n amed service is restarted.
By default, n amed sends standard messages to the rsyslo g daemon, which places them
in /var/lo g /messag es. Several standard channels are built into BIND with various severity
levels, such as d ef au lt _syslo g (which handles informational logging messages) and
d ef au lt _d eb u g (which specifically handles debugging messages). A default category,
called d ef au lt , uses the built-in channels to do normal logging without any special
configuration.
Customizing the logging process can be a very detailed process and is beyond the scope
of this chapter. For information on creating custom BIND logs, see the BIND 9 Administrator
Reference Manual referenced in Section 16.2.7.1, “ Installed D ocumentation” .
server
The server statement allows you to specify options that affect how the n amed service
should respond to remote nameservers, especially with regard to notifications and zone
transfers.

337

Deployment G uide

The t ran sf er- f o rmat option controls the number of resource records that are sent with
each message. It can be either o n e- an swer (only one resource record), or man yan swers (multiple resource records). Note that while the man y- an swers option is more
efficient, it is not supported by older versions of BIND .
t ru st ed - keys
The t ru st ed - keys statement allows you to specify assorted public keys used for secure
D NS (D NSSEC). See Section 16.2.5.4, “ D NS Security Extensions (D NSSEC)” for more
information on this topic.
view
The view statement allows you to create special views depending upon which network the
host querying the nameserver is on. This allows some hosts to receive one answer
regarding a zone while other hosts receive totally different information. Alternatively, certain
zones may only be made available to particular trusted hosts while non-trusted hosts can
only make queries for other zones.
Multiple views can be used as long as their names are unique. The mat ch - clien t s option
allows you to specify the IP addresses that apply to a particular view. If the o p t io n s
statement is used within a view, it overrides the already configured global options. Finally,
most view statements contain multiple z o n e statements that apply to the mat ch - clien t s
list.
Note that the order in which the view statements are listed is important, as the first statement
that matches a particular client's IP address is used. For more information on this topic, see
Section 16.2.5.1, “ Multiple Views” .

1 6 .2 .1 .3. Co m m e nt T ags
Additionally to statements, the /et c/n amed .co n f file can also contain comments. Comments are
ignored by the n amed service, but can prove useful when providing additional information to a user.
The following are valid comment tags:
//
Any text after the // characters to the end of the line is considered a comment. For example:
notify yes; // notify all secondary nameservers
#
Any text after the # character to the end of the line is considered a comment. For example:
notify yes; # notify all secondary nameservers
/* an d */
Any block of text enclosed in /* and */ is considered a comment. For example:
notify yes; /* notify all secondary nameservers */

16.2.2. Edit ing Zone Files
As outlined in Section 16.1.1, “ Nameserver Z ones” , zone files contain information about a

338

⁠Chapt er 1 6 . DNS Servers

namespace. They are stored in the n amed working directory located in /var/n amed / by default, and
each zone file is named according to the f ile option in the z o n e statement, usually in a way that
relates to the domain in question and identifies the file as containing zone data, such as
examp le.co m.z o n e.
T ab le 16 .5. T h e n amed service z o n e f iles
Pat h

D escrip t io n

/var/n amed /

The working directory for the n amed service.
The nameserver is not allowed to write to this
directory.
The directory for secondary zones. This
directory is writable by the n amed service.
The directory for other files, such as dynamic
D NS (D D NS) zones or managed D NSSEC keys.
This directory is writable by the n amed service.
The directory for various statistics and
debugging files. This directory is writable by the
n amed service.

/var/n amed /slaves/
/var/n amed /d yn amic/

/var/n amed /d at a/

A zone file consists of directives and resource records. D irectives tell the nameserver to perform tasks
or apply special settings to the zone, resource records define the parameters of the zone and assign
identities to individual hosts. While the directives are optional, the resource records are required in
order to provide name service to a zone.
All directives and resource records should be entered on individual lines.

1 6 .2 .2 .1 . Co m m o n Dire ct ive s
D irectives begin with the dollar sign character followed by the name of the directive, and usually
appear at the top of the file. The following directives are commonly used in zone files:
$IN C LU D E
The $IN C LU D E directive allows you to include another file at the place where it appears,
so that other zone settings can be stored in a separate zone file.

Examp le 16 .7. U sin g t h e $IN C LU D E d irect ive
$INCLUD E /var/named/penguin.example.com
$O R IG IN
The $O R IG IN directive allows you to append the domain name to unqualified records,
such as those with the host name only. Note that the use of this directive is not necessary if
the zone is specified in /et c/n amed .co n f , since the zone name is used by default.
In Example 16.8, “ Using the $ORIGIN directive” , any names used in resource records that
do not end in a trailing period are appended with examp le.co m.

Examp le 16 .8. U sin g t h e $O R IG IN d irect ive
$ORIGIN example.com.

339

Deployment G uide

$T T L
The $T T L directive allows you to set the default Time to Live (TTL) value for the zone, that is,
how long is a zone record valid. Each resource record can contain its own TTL value,
which overrides this directive.
Increasing this value allows remote nameservers to cache the zone information for a longer
period of time, reducing the number of queries for the zone and lengthening the amount of
time required to propagate resource record changes.

Examp le 16 .9 . U sin g t h e $T T L d irect ive
$TTL 1D

1 6 .2 .2 .2 . Co m m o n Re so urce Re co rds
The following resource records are commonly used in zone files:
A
The Address record specifies an IP address to be assigned to a name. It takes the following
form:
hostname IN A IP-address
If the hostname value is omitted, the record will point to the last specified hostname.
In Example 16.10, “ Using the A resource record” , the requests for server1.examp le.co m
are pointed to 10.0.1.3 or 10.0.1.5.

Examp le 16 .10. U sin g t h e A reso u rce reco rd
server1 IN A 10.0.1.3
IN A 10.0.1.5
C N AME
The Canonical Name record maps one name to another. Because of this, this type of record
is sometimes referred to as an alias record. It takes the following form:
alias-name IN CNAME real-name
C N AME records are most commonly used to point to services that use a common naming
scheme, such as www for Web servers. However, there are multiple restrictions for their
usage:
CNAME records should not point to other CNAME records. This is mainly to avoid
possible infinite loops.
CNAME records should not contain other resource record types (such as A, NS, MX,
etc.). The only exception are D NSSEC related records (that is, RRSIG, NSEC, etc.) when
the zone is signed.

34 0

⁠Chapt er 1 6 . DNS Servers

Other resource record that point to the fully qualified domain name (FQD N) of a host
(that is, NS, MX, PTR) should not point to a CNAME record.
In Example 16.11, “ Using the CNAME resource record” , the A record binds a host name to
an IP address, while the C N AME record points the commonly used www host name to it.

Examp le 16 .11. U sin g t h e C N AME reso u rce reco rd
server1 IN A
10.0.1.5
www
IN CNAME server1
MX
The Mail Exchange record specifies where the mail sent to a particular namespace controlled
by this zone should go. It takes the following form:
IN MX preference-value email-server-name
The email-server-name is a fully qualified domain name (FQD N). The preference-value allows
numerical ranking of the email servers for a namespace, giving preference to some email
systems over others. The MX resource record with the lowest preference-value is preferred
over the others. However, multiple email servers can possess the same value to distribute
email traffic evenly among them.
In Example 16.12, “ Using the MX resource record” , the first mail.examp le.co m email server
is preferred to the mail2.examp le.co m email server when receiving email destined for the
examp le.co m domain.

Examp le 16 .12. U sin g t h e MX reso u rce reco rd
example.com. IN MX 10 mail.example.com.
IN MX 20 mail2.example.com.
NS
The Nameserver record announces authoritative nameservers for a particular zone. It takes
the following form:
IN NS nameserver-name
The nameserver-name should be a fully qualified domain name (FQD N). Note that when two
nameservers are listed as authoritative for the domain, it is not important whether these
nameservers are secondary nameservers, or if one of them is a primary server. They are
both still considered authoritative.

Examp le 16 .13. U sin g t h e N S reso u rce reco rd
IN NS dns1.example.com.
IN NS dns2.example.com.
PT R
The Pointer record points to another part of the namespace. It takes the following form:

34 1

Deployment G uide

The Pointer record points to another part of the namespace. It takes the following form:
last-IP-digit IN PTR FQDN-of-system
The last-IP-digit directive is the last number in an IP address, and the FQDN-of-system is a
fully qualified domain name (FQD N).
PT R records are primarily used for reverse name resolution, as they point IP addresses
back to a particular name. See Section 16.2.2.4.2, “ A Reverse Name Resolution Z one File”
for more examples of PT R records in use.
SO A
The Start of Authority record announces important authoritative information about a
namespace to the nameserver. Located after the directives, it is the first resource record in a
zone file. It takes the following form:
@ IN SOA primary-name-server hostmaster-email (
serial-number
time-to-refresh
time-to-retry
time-to-expire
minimum-TTL )
The directives are as follows:
The @ symbol places the $O R IG IN directive (or the zone's name if the $O R IG IN
directive is not set) as the namespace being defined by this SO A resource record.
The primary-name-server directive is the host name of the primary nameserver that is
authoritative for this domain.
The hostmaster-email directive is the email of the person to contact about the namespace.
The serial-number directive is a numerical value incremented every time the zone file is
altered to indicate it is time for the n amed service to reload the zone.
The time-to-refresh directive is the numerical value secondary nameservers use to
determine how long to wait before asking the primary nameserver if any changes have
been made to the zone.
The time-to-retry directive is a numerical value used by secondary nameservers to
determine the length of time to wait before issuing a refresh request in the event that the
primary nameserver is not answering. If the primary server has not replied to a refresh
request before the amount of time specified in the time-to-expire directive elapses, the
secondary servers stop responding as an authority for requests concerning that
namespace.
In BIND 4 and 8, the minimum-TTL directive is the amount of time other nameservers
cache the zone's information. In BIND 9, it defines how long negative answers are
cached for. Caching of negative answers can be set to a maximum of 3 hours (that is,
3H ).
When configuring BIND , all times are specified in seconds. However, it is possible to use
abbreviations when specifying units of time other than seconds, such as minutes (M),
hours (H ), days (D ), and weeks (W). Table 16.6, “ Seconds compared to other time units”
shows an amount of time in seconds and the equivalent time in another format.

34 2

⁠Chapt er 1 6 . DNS Servers

T ab le 16 .6 . Seco n d s co mp ared t o o t h er t ime u n it s
Seco n d s

O t h er T ime U n it s

60
1800
3600
10800
21600
43200
86400
259200
604800
31536000

1M
30M
1H
3H
6H
12H
1D
3D
1W
36 5D

Examp le 16 .14 . U sin g t h e SO A reso u rce reco rd
@ IN SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 ) ; minimum TTL of 1 day

1 6 .2 .2 .3. Co m m e nt T ags
Additionally to resource records and directives, a zone file can also contain comments. Comments
are ignored by the n amed service, but can prove useful when providing additional information to the
user. Any text after the semicolon character to the end of the line is considered a comment. For
example:
604800 ; expire after 1 week

1 6 .2 .2 .4 . Exam ple Usage
The following examples show the basic usage of zone files.
16 .2.2.4 .1. A Simp le Z o n e File
Example 16.15, “ A simple zone file” demonstrates the use of standard directives and SO A values.

Examp le 16 .15. A simp le z o n e f ile
$ORIGIN example.com.
$TTL 86400
@
IN SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 ) ; minimum TTL of 1 day
;

34 3

Deployment G uide

;
IN NS dns1.example.com.
IN NS dns2.example.com.
dns1
IN A
10.0.1.1
IN AAAA aaaa:bbbb::1
dns2
IN A
10.0.1.2
IN AAAA aaaa:bbbb::2
;
;
@
IN MX 10 mail.example.com.
IN MX 20 mail2.example.com.
mail
IN A
10.0.1.5
IN AAAA aaaa:bbbb::5
mail2 IN A
10.0.1.6
IN AAAA aaaa:bbbb::6
;
;
; This sample zone file illustrates sharing the same IP addresses
; for multiple services:
;
services IN A
10.0.1.10
IN AAAA aaaa:bbbb::10
IN A
10.0.1.11
IN AAAA aaaa:bbbb::11
ftp
IN CNAME services.example.com.
www
IN CNAME services.example.com.
;
;

In this example, the authoritative nameservers are set as d n s1.examp le.co m and
d n s2.examp le.co m, and are tied to the 10.0.1.1 and 10.0.1.2 IP addresses respectively using the
A record.
The email servers configured with the MX records point to mail and mail2 via A records. Since these
names do not end in a trailing period, the $O R IG IN domain is placed after them, expanding them to
mail.examp le.co m and mail2.examp le.co m.
Services available at the standard names, such as www.examp le.co m (WWW), are pointed at the
appropriate servers using the C N AME record.
This zone file would be called into service with a z o n e statement in the /et c/n amed .co n f similar to
the following:
zone " example.com" IN {
type master;
file " example.com.zone" ;
allow-update { none; };
};
16 .2.2.4 .2. A R everse N ame R eso lu t io n Z o n e File

34 4

⁠Chapt er 1 6 . DNS Servers

A reverse name resolution zone file is used to translate an IP address in a particular namespace into
an fully qualified domain name (FQD N). It looks very similar to a standard zone file, except that the
PT R resource records are used to link the IP addresses to a fully qualified domain name as shown in
Example 16.16, “ A reverse name resolution zone file” .

Examp le 16 .16 . A reverse n ame reso lu t io n z o n e f ile
$ORIGIN 1.0.10.in-addr.arpa.
$TTL 86400
@ IN SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 ) ; minimum TTL of 1 day
;
@ IN NS dns1.example.com.
;
1 IN PTR dns1.example.com.
2 IN PTR dns2.example.com.
;
5 IN PTR server1.example.com.
6 IN PTR server2.example.com.
;
3 IN PTR ftp.example.com.
4 IN PTR ftp.example.com.

In this example, IP addresses 10.0.1.1 through 10.0.1.6 are pointed to the corresponding fully
qualified domain name.
This zone file would be called into service with a z o n e statement in the /et c/n amed .co n f file similar
to the following:
zone " 1.0.10.in-addr.arpa" IN {
type master;
file " example.com.rr.zone" ;
allow-update { none; };
};
There is very little difference between this example and a standard z o n e statement, except for the
zone name. Note that a reverse name resolution zone requires the first three blocks of the IP address
reversed followed by .in - ad d r.arp a. This allows the single block of IP numbers used in the reverse
name resolution zone file to be associated with the zone.

16.2.3. Using t he rndc Ut ilit y
The rn d c utility is a command-line tool that allows you to administer the n amed service, both locally
and from a remote machine. Its usage is as follows:
rn d c [option...] command [command-option]

1 6 .2 .3.1 . Co nfiguring t he Ut ilit y

34 5

Deployment G uide

To prevent unauthorized access to the service, n amed must be configured to listen on the selected
port (that is, 9 53 by default), and an identical key must be used by both the service and the rn d c
utility.
T ab le 16 .7. R elevan t f iles
Pat h

D escrip t io n

/et c/n amed .co n f
/et c/rn d c.co n f
/et c/rn d c.key

The default configuration file for the n amed service.
The default configuration file for the rn d c utility.
The default key location.

The rn d c configuration is located in /et c/rn d c.co n f . If the file does not exist, the utility will use the
key located in /et c/rn d c.key, which was generated automatically during the installation process
using the rn d c- co n f g en - a command.
The n amed service is configured using the co n t ro ls statement in the /et c/n amed .co n f
configuration file as described in Section 16.2.1.2, “ Other Statement Types” . Unless this statement is
present, only the connections from the loopback address (that is, 127.0.0.1) will be allowed, and the
key located in /et c/rn d c.key will be used.
For more information on this topic, see manual pages and the BIND 9 Administrator Reference Manual
listed in Section 16.2.7, “ Additional Resources” .

Set the correct permissions
To prevent unprivileged users from sending control commands to the service, make sure only
root is allowed to read the /et c/rn d c.key file:
~]# ch mo d o - rwx /et c/rn d c.key

1 6 .2 .3.2 . Che cking t he Se rvice St at us
To check the current status of the n amed service, use the following command:
~]# rn d c st at u s
version: 9.7.0-P2-RedHat-9.7.0-5.P2.el6
CPUs found: 1
worker threads: 1
number of zones: 16
debug level: 0
xfers running: 0
xfers deferred: 0
soa queries in progress: 0
query logging is OFF
recursive clients: 0/0/1000
tcp clients: 0/100
server is up and running

1 6 .2 .3.3. Re lo ading t he Co nfigurat io n and Zo ne s
To reload both the configuration file and zones, type the following at a shell prompt:

34 6

⁠Chapt er 1 6 . DNS Servers

~]# rn d c relo ad
server reload successful
This will reload the zones while keeping all previously cached responses, so that you can make
changes to the zone files without losing all stored name resolutions.
To reload a single zone, specify its name after the relo ad command, for example:
~]# rn d c relo ad lo calh o st
zone reload up-to-date
Finally, to reload the configuration file and newly added zones only, type:
~]# rn d c reco n f ig

Modifying zones with dynamic DNS
If you intend to manually modify a zone that uses D ynamic D NS (D D NS), make sure you run
the f reez e command first:
~]# rn d c f reez e lo calh o st
Once you are finished, run the t h aw command to allow the D D NS again and reload the zone:
~]# rn d c t h aw lo calh o st
The zone reload and thaw was successful.

1 6 .2 .3.4 . Updat ing Zo ne Ke ys
To update the D NSSEC keys and sign the zone, use the sig n command. For example:
~]# rn d c sig n lo calh o st
Note that to sign a zone with the above command, the au t o - d n ssec option has to be set to
main t ain in the zone statement. For instance:
zone " localhost" IN {
type master;
file " named.localhost" ;
allow-update { none; };
auto-dnssec maintain;
};

1 6 .2 .3.5 . Enabling t he DNSSEC Validat io n
To enable the D NSSEC validation, type the following at a shell prompt:
~]# rn d c valid at io n o n

34 7

Deployment G uide

Similarly, to disable this option, type:
~]# rn d c valid at io n o f f
See the o p t io n s statement described in Section 16.2.1.1, “ Common Statement Types” for information
on how to configure this option in /et c/n amed .co n f .

1 6 .2 .3.6 . Enabling t he Que ry Lo gging
To enable (or disable in case it is currently enabled) the query logging, run the following command:
~]# rn d c q u erylo g
To check the current setting, use the st at u s command as described in Section 16.2.3.2, “ Checking
the Service Status” .

16.2.4 . Using t he dig Ut ilit y
The d ig utility is a command-line tool that allows you to perform D NS lookups and debug a
nameserver configuration. Its typical usage is as follows:
d ig [@server] [option...] name type
See Section 16.2.2.2, “ Common Resource Records” for a list of common types.

1 6 .2 .4 .1 . Lo o king Up a Nam e se rve r
To look up a nameserver for a particular domain, use the command in the following form:
d ig name NS
In Example 16.17, “ A sample nameserver lookup” , the d ig utility is used to display nameservers for
examp le.co m.

Examp le 16 .17. A samp le n ameserver lo o ku p
~]$ d ig examp le.co m N S
; <<>> D iG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> example.com NS
;; global options: +cmd
;; Got answer:
;; ->>HEAD ER<<- opcode: QUERY, status: NOERROR, id: 57883
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, AD D ITIONAL: 0
;; QUESTION SECTION:
;example.com.
IN
;; ANSWER SECTION:
example.com.
99374 IN
example.com.
99374 IN

NS

NS
NS

;; Query time: 1 msec
;; SERVER: 10.34.255.7#53(10.34.255.7)

34 8

a.iana-servers.net.
b.iana-servers.net.

⁠Chapt er 1 6 . DNS Servers

;; WHEN: Wed Aug 18 18:04:06 2010
;; MSG SIZ E rcvd: 77

1 6 .2 .4 .2 . Lo o king Up an IP Addre ss
To look up an IP address assigned to a particular domain, use the command in the following form:
d ig name A
In Example 16.18, “ A sample IP address lookup” , the d ig utility is used to display the IP address of
examp le.co m.

Examp le 16 .18. A samp le IP ad d ress lo o ku p
~]$ d ig examp le.co m A
; <<>> D iG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> example.com A
;; global options: +cmd
;; Got answer:
;; ->>HEAD ER<<- opcode: QUERY, status: NOERROR, id: 4849
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, AD D ITIONAL: 0
;; QUESTION SECTION:
;example.com.
IN

A

;; ANSWER SECTION:
example.com.
155606 IN

A

;; AUTHORITY SECTION:
example.com.
99175 IN
example.com.
99175 IN

NS
NS

192.0.32.10

a.iana-servers.net.
b.iana-servers.net.

;; Query time: 1 msec
;; SERVER: 10.34.255.7#53(10.34.255.7)
;; WHEN: Wed Aug 18 18:07:25 2010
;; MSG SIZ E rcvd: 93

1 6 .2 .4 .3. Lo o king Up a Ho st nam e
To look up a host name for a particular IP address, use the command in the following form:
d ig - x address
In Example 16.19, “ A sample host name lookup” , the d ig utility is used to display the host name
assigned to 19 2.0.32.10.

Examp le 16 .19 . A samp le h o st n ame lo o ku p
~]$ d ig - x 19 2.0.32.10

34 9

Deployment G uide

; <<>> D iG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> -x 192.0.32.10
;; global options: +cmd
;; Got answer:
;; ->>HEAD ER<<- opcode: QUERY, status: NOERROR, id: 29683
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 5, AD D ITIONAL: 6
;; QUESTION SECTION:
;10.32.0.192.in-addr.arpa.

IN

PTR

;; ANSWER SECTION:
10.32.0.192.in-addr.arpa. 21600 IN
;; AUTHORITY SECTION:
32.0.192.in-addr.arpa. 21600
32.0.192.in-addr.arpa. 21600
32.0.192.in-addr.arpa. 21600
32.0.192.in-addr.arpa. 21600
32.0.192.in-addr.arpa. 21600

IN
IN
IN
IN
IN

PTR

NS
NS
NS
NS
NS

www.example.com.

b.iana-servers.org.
c.iana-servers.net.
d.iana-servers.net.
ns.icann.org.
a.iana-servers.net.

;; AD D ITIONAL SECTION:
a.iana-servers.net. 13688 IN
A
192.0.34.43
b.iana-servers.org. 5844 IN
A
193.0.0.236
b.iana-servers.org. 5844 IN
AAAA 2001:610:240:2::c100:ec
c.iana-servers.net. 12173 IN
A
139.91.1.10
c.iana-servers.net. 12173 IN
AAAA 2001:648:2c30::1:10
ns.icann.org.
12884 IN
A
192.0.34.126
;; Query time: 156 msec
;; SERVER: 10.34.255.7#53(10.34.255.7)
;; WHEN: Wed Aug 18 18:25:15 2010
;; MSG SIZ E rcvd: 310

16.2.5. Advanced Feat ures of BIND
Most BIND implementations only use the n amed service to provide name resolution services or to act
as an authority for a particular domain. However, BIND version 9 has a number of advanced features
that allow for a more secure and efficient D NS service.

Make sure the feature is supported
Before attempting to use advanced features like D NSSEC, TSIG, or IXFR (Incremental Z one
Transfer), make sure that the particular feature is supported by all nameservers in the network
environment, especially when you use older versions of BIND or non-BIND servers.
All of the features mentioned are discussed in greater detail in the BIND 9 Administrator Reference
Manual referenced in Section 16.2.7.1, “ Installed D ocumentation” .

1 6 .2 .5 .1 . Mult iple Vie ws
Optionally, different information can be presented to a client depending on the network a request
originates from. This is primarily used to deny sensitive D NS entries from clients outside of the local
network, while allowing queries from clients inside the local network.

350

⁠Chapt er 1 6 . DNS Servers

To configure multiple views, add the view statement to the /et c/n amed .co n f configuration file. Use
the mat ch - clien t s option to match IP addresses or entire networks and give them special options
and zone data.

1 6 .2 .5 .2 . Incre m e nt al Zo ne T ransfe rs (IXFR)
Incremental Zone Transfers (IXFR) allow a secondary nameserver to only download the updated
portions of a zone modified on a primary nameserver. Compared to the standard transfer process,
this makes the notification and update process much more efficient.
Note that IXFR is only available when using dynamic updating to make changes to master zone
records. If manually editing zone files to make changes, Automatic Zone Transfer (AXFR) is used.

1 6 .2 .5 .3. T ransact io n SIGnat ure s (T SIG)
Transaction SIGnatures (TSIG) ensure that a shared secret key exists on both primary and secondary
nameserver before allowing a transfer. This strengthens the standard IP address-based method of
transfer authorization, since attackers would not only need to have access to the IP address to
transfer the zone, but they would also need to know the secret key.
Since version 9, BIND also supports TKEY, which is another shared secret key method of authorizing
zone transfers.

Secure the transfer
When communicating over an insecure network, do not rely on IP address-based
authentication only.

1 6 .2 .5 .4 . DNS Se curit y Ext e nsio ns (DNSSEC)
Domain Name System Security Extensions (DNSSEC) provide origin authentication of D NS data,
authenticated denial of existence, and data integrity. When a particular domain is marked as secure,
the SER FVAIL response is returned for each resource record that fails the validation.
Note that to debug a D NSSEC-signed domain or a D NSSEC-aware resolver, you can use the d ig
utility as described in Section 16.2.4, “ Using the dig Utility” . Useful options are + d n ssec (requests
D NSSEC-related resource records by setting the D NSSEC OK bit), + cd (tells recursive nameserver
not to validate the response), and + b u f siz e= 512 (changes the packet size to 512B to get through
some firewalls).

1 6 .2 .5 .5 . Int e rne t Pro t o co l ve rsio n 6 (IPv6 )
Internet Protocol version 6 (IPv6) is supported through the use of AAAA resource records, and the
list en - o n - v6 directive as described in Table 16.3, “ Commonly used options” .

16.2.6. Common Mist akes t o Avoid
The following is a list of recommendations on how to avoid common mistakes users make when
configuring a nameserver:
U se semico lo n s an d cu rly b racket s co rrect ly
An omitted semicolon or unmatched curly bracket in the /et c/n amed .co n f file can prevent
the n amed service from starting.

351

Deployment G uide

U se p erio d co rrect ly
In zone files, a period at the end of a domain name denotes a fully qualified domain name.
If omitted, the n amed service will append the name of the zone or the value of $O R IG IN to
complete it.
In cremen t t h e serial n u mb er wh en ed it in g a z o n e f ile
If the serial number is not incremented, the primary nameserver will have the correct, new
information, but the secondary nameservers will never be notified of the change, and will
not attempt to refresh their data of that zone.
C o n f ig u re t h e f irewall
If a firewall is blocking connections from the n amed service to other nameservers, the
recommended practice is to change the firewall settings.

Avoid using fixed UDP source ports
According to the recent research in D NS security, using a fixed UD P source port for
D NS queries is a potential security vulnerability that could allow an attacker to
conduct cache-poisoning attacks more easily. To prevent this, configure your
firewall to allow queries from a random UD P source port.

16.2.7. Addit ional Resources
The following sources of information provide additional resources regarding BIND .

1 6 .2 .7 .1 . Inst alle d Do cum e nt at io n
BIND features a full range of installed documentation covering many different topics, each placed in
its own subject directory. For each item below, replace version with the version of the bind package
installed on the system:
/u sr/sh are/d o c/b in d - version/
The main directory containing the most recent documentation.
/u sr/sh are/d o c/b in d - version/arm/
The directory containing the BIND 9 Administrator Reference Manual in HTML and SGML
formats, which details BIND resource requirements, how to configure different types of
nameservers, how to perform load balancing, and other advanced topics. For most new
users of BIND , this is the best place to start.
/u sr/sh are/d o c/b in d - version/d raf t /
The directory containing assorted technical documents that review issues related to the
D NS service, and propose some methods to address them.
/u sr/sh are/d o c/b in d - version/misc/
The directory designed to address specific advanced issues. Users of BIND version 8
should consult the mig rat io n document for specific changes they must make when moving
to BIND 9. The o p t io n s file lists all of the options implemented in BIND 9 that are used in
/et c/n amed .co n f .

352

⁠Chapt er 1 6 . DNS Servers

/u sr/sh are/d o c/b in d - version/rf c/
The directory providing every RFC document related to BIND .
There is also a number of man pages for the various applications and configuration files involved
with BIND :
man rn d c
The manual page for rn d c containing the full documentation on its usage.
man n amed
The manual page for n amed containing the documentation on assorted arguments that
can be used to control the BIND nameserver daemon.
man lwresd
The manual page for lwresd containing the full documentation on the lightweight resolver
daemon and its usage.
man n amed .co n f
The manual page with a comprehensive list of options available within the n amed
configuration file.
man rn d c.co n f
The manual page with a comprehensive list of options available within the rn d c
configuration file.

1 6 .2 .7 .2 . Use ful We bsit e s
h t t p ://www.isc.o rg /so f t ware/b in d
The home page of the BIND project containing information about current releases as well
as a PD F version of the BIND 9 Administrator Reference Manual.

1 6 .2 .7 .3. Re lat e d Bo o ks
DNS and BIND b y Pau l Alb it z an d C ricket Liu ; O ' R eilly & Asso ciat es
A popular reference that explains both common and esoteric BIND configuration options,
and provides strategies for securing a D NS server.
The Concise Guide to DNS and BIND b y N ico lai Lan g f eld t ; Q u e
Looks at the connection between multiple network services and BIND , with an emphasis on
task-oriented, technical topics.

353

Deployment G uide

Chapter 17. Web Servers
H T T P (Hypertext Transfer Protocol) server, or a web server, is a network service that serves content
to a client over the web. This typically means web pages, but any other documents can be served as
well.

17.1. T he Apache HT T P Server
This section focuses on the Ap ach e H T T P Server 2.2, a robust, full-featured open source web
server developed by the Apache Software Foundation, that is included in Red Hat Enterprise Linux 6.
It describes the basic configuration of the h t t p d service, and covers advanced topics such as
adding server modules, setting up virtual hosts, or configuring the secure HTTP server.
There are important differences between the Apache HTTP Server 2.2 and version 2.0, and if you are
upgrading from a previous release of Red Hat Enterprise Linux, you will need to update the h t t p d
service configuration accordingly. This section reviews some of the newly added features, outlines
important changes, and guides you through the update of older configuration files.

17.1.1. New Feat ures
The Apache HTTP Server version 2.2 introduces the following enhancements:
Improved caching modules, that is, mo d _cach e and mo d _d isk_cach e.
Support for proxy load balancing, that is, the mo d _p ro xy_b alan cer module.
Support for large files on 32-bit architectures, allowing the web server to handle files greater than
2GB.
A new structure for authentication and authorization support, replacing the authentication
modules provided in previous versions.

17.1.2. Not able Changes
Since version 2.0, few changes have been made to the default h t t p d service configuration:
The following modules are no longer loaded by default: mo d _cern _met a and mo d _asis.
The following module is newly loaded by default: mo d _ext _f ilt er.

17.1.3. Updat ing t he Configurat ion
To update the configuration files from the Apache HTTP Server version 2.0, take the following steps:
1. Make sure all module names are correct, since they may have changed. Adjust the
Lo ad Mo d u le directive for each module that has been renamed.
2. Recompile all third party modules before attempting to load them. This typically means
authentication and authorization modules.
3. If you use the mo d _u serd ir module, make sure the U serD ir directive indicating a directory
name (typically p u b lic_h t ml) is provided.
4. If you use the Apache HTTP Secure Server, see Section 17.1.9, “ Enabling the mod_ssl
Module” for important information on enabling the Secure Sockets Layer (SSL) protocol.

354

⁠Chapt er 1 7 . Web Servers

Note that you can check the configuration for possible errors by using the following command:
~]# service h t t p d co n f ig t est
Syntax OK
For more information on upgrading the Apache HTTP Server configuration from version 2.0 to 2.2,
see http://httpd.apache.org/docs/2.2/upgrading.html.

17.1.4 . Running t he ht t pd Service
This section describes how to start, stop, restart, and check the current status of the Apache HTTP
Server. To be able to use the h t t p d service, make sure you have the httpd installed. You can do so
by using the following command:
~]# yu m in st all h t t p d
For more information on the concept of runlevels and how to manage system services in Red Hat
Enterprise Linux in general, see Chapter 11, Services and Daemons.

1 7 .1 .4 .1 . St art ing t he Se rvice
To run the h t t p d service, type the following at a shell prompt as ro o t :
~]# service h t t p d st art
Starting httpd:

[ OK ]

If you want the service to start automatically at the boot time, use the following command:
~]# ch kco n f ig h t t p d o n
This will enable the service for runlevel 2, 3, 4, and 5. Alternatively, you can use the Service
C o n f ig u rat io n utility as described in Section 11.2.1.1, “ Enabling and D isabling a Service” .

Using the secure server
If running the Apache HTTP Server as a secure server, a password may be required after the
machine boots if using an encrypted private SSL key.

1 7 .1 .4 .2 . St o pping t he Se rvice
To stop the running h t t p d service, type the following at a shell prompt as ro o t :
~]# service h t t p d st o p
Stopping httpd:

[ OK ]

To prevent the service from starting automatically at the boot time, type:
~]# ch kco n f ig h t t p d o f f

355

Deployment G uide

This will disable the service for all runlevels. Alternatively, you can use the Service C o n f ig u rat io n
utility as described in Section 11.2.1.1, “ Enabling and D isabling a Service” .

1 7 .1 .4 .3. Re st art ing t he Se rvice
There are three different ways to restart a running h t t p d service:
1. To restart the service completely, enter the following command as ro o t :
~]# service h t t p d rest art
Stopping httpd:
Starting httpd:

[ OK ]
[ OK ]

This stops the running h t t p d service and immediately starts it again. Use this command after
installing or removing a dynamically loaded module such as PHP.
2. To only reload the configuration, as ro o t , type:
~]# service h t t p d relo ad
This causes the running h t t p d service to reload its configuration file. Any requests being
currently processed will be interrupted, which may cause a client browser to display an error
message or render a partial page.
3. To reload the configuration without affecting active requests, enter the following command as
ro o t :
~]# service h t t p d g racef u l
This causes the running h t t p d service to reload its configuration file. Any requests being
currently processed will use the old configuration.
Alternatively, you can use the Service C o n f ig u rat io n utility as described in Section 11.2.1.2,
“ Starting, Restarting, and Stopping a Service” .

1 7 .1 .4 .4 . Ve rifying t he Se rvice St at us
To verify that the h t t p d service is running, type the following at a shell prompt:
~]# service h t t p d st at u s
httpd (pid 19014) is running...
Alternatively, you can use the Service C o n f ig u rat io n utility as described in Section 11.2.1, “ Using
the Service Configuration Utility” .

17.1.5. Edit ing t he Configurat ion Files
When the h t t p d service is started, by default, it reads the configuration from locations that are listed
in Table 17.1, “ The httpd service configuration files” .
T ab le 17.1. T h e h t t p d service co n f ig u rat io n f iles
Pat h

D escrip t io n

/et c/h t t p d /co n f /h t t p d .co n f

The main configuration file.

356

⁠Chapt er 1 7 . Web Servers

Pat h

D escrip t io n

/et c/h t t p d /co n f .d /

An auxiliary directory for configuration files that are included in
the main configuration file.

Although the default configuration should be suitable for most situations, it is a good idea to become
at least familiar with some of the more important configuration options. Note that for any changes to
take effect, the web server has to be restarted first. See Section 17.1.4.3, “ Restarting the Service” for
more information on how to restart the h t t p d service.
To check the configuration for possible errors, type the following at a shell prompt:
~]# service h t t p d co n f ig t est
Syntax OK
To make the recovery from mistakes easier, it is recommended that you make a copy of the original
file before editing it.

1 7 .1 .5 .1 . Co m m o n ht t pd.co nf Dire ct ive s
The following directives are commonly used in the /et c/h t t p d /co n f /h t t p d .co n f configuration file:

The  directive allows you to apply certain directives to a particular directory
only. It takes the following form:

directive
…

The directory can be either a full path to an existing directory in the local file system, or a
wildcard expression.
This directive can be used to configure additional cg i- b in directories for server-side
scripts located outside the directory that is specified by Scrip t Alias. In this case, the
ExecC G I and Ad d H an d ler directives must be supplied, and the permissions on the target
directory must be set correctly (that is, 0755).

Examp le 17.1. U sin g t h e  d irect ive

Options Indexes FollowSymLinks
AllowOverride None
Order allow,deny
Allow from all


The If D ef in e directive allows you to use certain directives only when a particular
parameter is supplied on the command line. It takes the following form:

357

Deployment G uide


directive
…

The parameter can be supplied at a shell prompt using the - D parameter command-line
option (for example, h t t p d - D En ab leH o me). If the optional exclamation mark (that is, !) is
present, the enclosed directives are used only when the parameter is not specified.

Examp le 17.2. U sin g t h e  d irect ive

UserD ir public_html


The  directive allows you to use certain directive only when a particular
module is loaded. It takes the following form:

directive
…

The module can be identified either by its name, or by the file name. If the optional
exclamation mark (that is, !) is present, the enclosed directives are used only when the
module is not loaded.

Examp le 17.3. U sin g t h e  d irect ive

CacheEnable disk /
CacheRoot /var/cache/mod_proxy


The  directive allows you to apply certain directives to a particular URL only. It
takes the following form:

directive
…

The url can be either a path relative to the directory specified by the D o cu men t R o o t
directive (for example, /server- in f o ), or an external URL such as
h t t p ://examp le.co m/server- in f o .

Examp le 17.4 . U sin g t h e  d irect ive

358

⁠Chapt er 1 7 . Web Servers


SetHandler server-info
Order deny,allow
D eny from all
Allow from .example.com


The  directive allows you to apply certain directives to the proxy server only. It
takes the following form:

directive
…

The pattern can be an external URL, or a wildcard expression (for example,
h t t p ://examp le.co m/*).

Examp le 17.5. U sin g t h e  d irect ive

Order deny,allow
D eny from all
Allow from .example.com


The  directive allows you apply certain directives to particular virtual hosts
only. It takes the following form:

directive
…

The address can be an IP address, a fully qualified domain name, or a special form as
described in Table 17.2, “ Available  options” .
T ab le 17.2. Availab le  o p t io n s
O p t io n

D escrip t io n

*
_d ef au lt _

Represents all IP addresses.
Represents unmatched IP addresses.

Examp le 17.6 . U sin g t h e  d irect ive

ServerAdmin webmaster@penguin.example.com
D ocumentRoot /www/docs/penguin.example.com

359

Deployment G uide

ServerName penguin.example.com
ErrorLog logs/penguin.example.com-error_log
CustomLog logs/penguin.example.com-access_log common

AccessFileN ame
The AccessFileN ame directive allows you to specify the file to be used to customize
access control information for each directory. It takes the following form:
AccessFileName filename…
The filename is a name of the file to look for in the requested directory. By default, the server
looks for .h t access.
For security reasons, the directive is typically followed by the Files tag to prevent the files
beginning with .h t from being accessed by web clients. This includes the .h t access and
.h t p asswd files.

Examp le 17.7. U sin g t h e AccessFileN ame d irect ive
AccessFileName .htaccess

Order allow,deny
D eny from all
Satisfy All

Act io n
The Act io n directive allows you to specify a CGI script to be executed when a certain
media type is requested. It takes the following form:
Action content-type path
The content-type has to be a valid MIME type such as t ext /h t ml, imag e/p n g , or
ap p licat io n /p d f . The path refers to an existing CGI script, and must be relative to the
directory specified by the D o cu men t R o o t directive (for example, /cg i- b in /p ro cessimag e.cg i).

Examp le 17.8. U sin g t h e Act io n d irect ive
Action image/png /cgi-bin/process-image.cgi
Ad d D escrip t io n
The Ad d D escrip t io n directive allows you to specify a short description to be displayed in
server-generated directory listings for a given file. It takes the following form:
AddD escription " description" filename…
The description should be a short text enclosed in double quotes (that is, "). The filename

360

⁠Chapt er 1 7 . Web Servers

can be a full file name, a file extension, or a wildcard expression.

Examp le 17.9 . U sin g t h e Ad d D escrip t io n d irect ive
AddD escription " GZ IP compressed tar archive" .tgz
Ad d En co d in g
The Ad d En co d in g directive allows you to specify an encoding type for a particular file
extension. It takes the following form:
AddEncoding encoding extension…
The encoding has to be a valid MIME encoding such as x- co mp ress, x- g z ip , etc. The
extension is a case sensitive file extension, and is conventionally written with a leading dot
(for example, .g z ).
This directive is typically used to instruct web browsers to decompress certain file types as
they are downloaded.

Examp le 17.10. U sin g t h e Ad d En co d in g d irect ive
AddEncoding x-gzip .gz .tgz
Ad d H an d ler
The Ad d H an d ler directive allows you to map certain file extensions to a selected handler.
It takes the following form:
AddHandler handler extension…
The handler has to be a name of a previously defined handler. The extension is a case
sensitive file extension, and is conventionally written with a leading dot (for example, .cg i).
This directive is typically used to treat files with the .cg i extension as CGI scripts regardless
of the directory they are in. Additionally, it is also commonly used to process server-parsed
HTML and image-map files.

Examp le 17.11. U sin g t h e Ad d H an d ler o p t io n
AddHandler cgi-script .cgi
Ad d Ico n
The Ad d Ico n directive allows you to specify an icon to be displayed for a particular file in
server-generated directory listings. It takes the following form:
AddIcon path pattern…
The path refers to an existing icon file, and must be relative to the directory specified by the
D o cu men t R o o t directive (for example, /ico n s/f o ld er.p n g ). The pattern can be a file
name, a file extension, a wildcard expression, or a special form as described in the

361

Deployment G uide

following table:
T ab le 17.3. Availab le Ad d Ico n o p t io n s
O p t io n

D escrip t io n

^^D IR EC T O R Y^^
^^B LAN K IC O N ^^

Represents a directory.
Represents a blank line.

Examp le 17.12. U sin g t h e Ad d Ico n d irect ive
AddIcon /icons/text.png .txt READ ME
Ad d Ico n B yEn co d in g
The Ad d Ico n B yEn co d in g directive allows you to specify an icon to be displayed for a
particular encoding type in server-generated directory listings. It takes the following form:
AddIconByEncoding path encoding…
The path refers to an existing icon file, and must be relative to the directory specified by the
D o cu men t R o o t directive (for example, /ico n s/co mp ressed .p n g ). The encoding has to
be a valid MIME encoding such as x- co mp ress, x- g z ip , etc.

Examp le 17.13. U sin g t h e Ad d Ico n B yEn co d in g d irect ive
AddIconByEncoding /icons/compressed.png x-compress x-gzip
Ad d Ico n B yT yp e
The Ad d Ico n B yT yp e directive allows you to specify an icon to be displayed for a
particular media type in server-generated directory listings. It takes the following form:
AddIconByType path content-type…
The path refers to an existing icon file, and must be relative to the directory specified by the
D o cu men t R o o t directive (for example, /ico n s/t ext .p n g ). The content-type has to be
either a valid MIME type (for example, t ext /h t ml or imag e/p n g ), or a wildcard expression
such as t ext /*, imag e/*, etc.

Examp le 17.14 . U sin g t h e Ad d Ico n B yT yp e d irect ive
AddIconByType /icons/video.png video/*
Ad d Lan g u ag e
The Ad d Lan g u ag e directive allows you to associate a file extension with a specific
language. It takes the following form:
AddLanguage language extension…

362

⁠Chapt er 1 7 . Web Servers

The language has to be a valid MIME language such as cs, en , or f r. The extension is a
case sensitive file extension, and is conventionally written with a leading dot (for example,
.cs).
This directive is especially useful for web servers that serve content in multiple languages
based on the client's language settings.

Examp le 17.15. U sin g t h e Ad d Lan g u ag e d irect ive
AddLanguage cs .cs .cz
Ad d T yp e
The Ad d T yp e directive allows you to define or override the media type for a particular file
extension. It takes the following form:
AddType content-type extension…
The content-type has to be a valid MIME type such as t ext /h t ml, imag e/p n g , etc. The
extension is a case sensitive file extension, and is conventionally written with a leading dot
(for example, .cs).

Examp le 17.16 . U sin g t h e Ad d T yp e d irect ive
AddType application/x-gzip .gz .tgz
Alias
The Alias directive allows you to refer to files and directories outside the default directory
specified by the D o cu men t R o o t directive. It takes the following form:
Alias url-path real-path
The url-path must be relative to the directory specified by the D o cu men t R o o t directive (for
example, /imag es/). The real-path is a full path to a file or directory in the local file system.
This directive is typically followed by the D irect o ry tag with additional permissions to
access the target directory. By default, the /ico n s/ alias is created so that the icons from
/var/www/ico n s/ are displayed in server-generated directory listings.

Examp le 17.17. U sin g t h e Alias d irect ive
Alias /icons/ /var/www/icons/

Options Indexes MultiViews FollowSymLinks
AllowOverride None
Order allow,deny
Allow from all

Allo w

363

Deployment G uide

The Allo w directive allows you to specify which clients have permission to access a given
directory. It takes the following form:
Allow from client…
The client can be a domain name, an IP address (both full and partial), a network/netmask
pair, or all for all clients.

Examp le 17.18. U sin g t h e Allo w d irect ive
Allow from 192.168.1.0/255.255.255.0
Allo wO verrid e
The Allo wO verrid e directive allows you to specify which directives in a .h t access file can
override the default configuration. It takes the following form:
AllowOverride type…
The type has to be one of the available grouping options as described in Table 17.4,
“ Available AllowOverride options” .
T ab le 17.4 . Availab le Allo wO verrid e o p t io n s
O p t io n

D escrip t io n

All

All directives in .h t access are allowed to override earlier
configuration settings.
None
No directive in .h t access is allowed to override earlier
configuration settings.
Au t h C o n f ig
Allows the use of authorization directives such as Au t h N ame,
Au t h T yp e, or R eq u ire.
FileIn f o
Allows the use of file type, metadata, and mo d _rewrit e directives
such as D ef au lt T yp e, R eq u est H ead er, or R ewrit eEn g in e, as
well as the Act io n directive.
In d exes
Allows the use of directory indexing directives such as
Ad d D escrip t io n , Ad d Ico n , or Fan cyIn d exin g .
Limit
Allows the use of host access directives, that is, Allo w, D en y, and
O rd er.
O p t io n s [=option,…] Allows the use of the O p t io n s directive. Additionally, you can
provide a comma-separated list of options to customize which
options can be set using this directive.
Examp le 17.19 . U sin g t h e Allo wO verrid e d irect ive
AllowOverride FileInfo AuthConfig Limit
B ro wserMat ch
The B ro wserMat ch directive allows you to modify the server behavior based on the
client's web browser type. It takes the following form:
BrowserMatch pattern variable…

364

⁠Chapt er 1 7 . Web Servers

The pattern is a regular expression to match the User-Agent HTTP header field. The variable
is an environment variable that is set when the header field matches the pattern.
By default, this directive is used to deny connections to specific browsers with known
issues, and to disable keepalives and HTTP header flushes for browsers that are known to
have problems with these actions.

Examp le 17.20. U sin g t h e B ro wserMat ch d irect ive
BrowserMatch " Mozilla/2" nokeepalive
C ach eD ef au lt Exp ire
The C ach eD ef au lt Exp ire option allows you to set how long to cache a document that
does not have any expiration date or the date of its last modification specified. It takes the
following form:
CacheD efaultExpire time
The time is specified in seconds. The default option is 36 00 (that is, one hour).

Examp le 17.21. U sin g t h e C ach eD ef au lt Exp ire d irect ive
CacheD efaultExpire 3600
C ach eD isab le
The C ach eD isab le directive allows you to disable caching of certain URLs. It takes the
following form:
CacheD isable path
The path must be relative to the directory specified by the D o cu men t R o o t directive (for
example, /f iles/).

Examp le 17.22. U sin g t h e C ach eD isab le d irect ive
CacheD isable /temporary
C ach eEn ab le
The C ach eEn ab le directive allows you to specify a cache type to be used for certain URLs.
It takes the following form:
CacheEnable type url
The type has to be a valid cache type as described in Table 17.5, “ Available cache types” .
The url can be a path relative to the directory specified by the D o cu men t R o o t directive (for
example, /imag es/), a protocol (for example, f t p ://), or an external URL such as
h t t p ://examp le.co m/.

365

Deployment G uide

T ab le 17.5. Availab le cach e t yp es
T yp e

D escrip t io n

mem
d isk
fd

The memory-based storage manager.
The disk-based storage manager.
The file descriptor cache.

Examp le 17.23. U sin g t h e C ach eEn ab le d irect ive
CacheEnable disk /
C ach eLast Mo d if ied Fact o r
The C ach eLast Mo d if ied Fact o r directive allows you to customize how long to cache a
document that does not have any expiration date specified, but that provides information
about the date of its last modification. It takes the following form:
CacheLastModifiedFactor number
The number is a coefficient to be used to multiply the time that passed since the last
modification of the document. The default option is 0.1 (that is, one tenth).

Examp le 17.24 . U sin g t h e C ach eLast Mo d if ied Fact o r d irect ive
CacheLastModifiedFactor 0.1
C ach eMaxExp ire
The C ach eMaxExp ire directive allows you to specify the maximum amount of time to
cache a document. It takes the following form:
CacheMaxExpire time
The time is specified in seconds. The default option is 86 4 00 (that is, one day).

Examp le 17.25. U sin g t h e C ach eMaxExp ire d irect ive
CacheMaxExpire 86400
C ach eN eg o t iat ed D o cs
The C ach eN eg o t iat ed D o cs directive allows you to enable caching of the documents that
were negotiated on the basis of content. It takes the following form:
CacheNegotiatedD ocs option
The option has to be a valid keyword as described in Table 17.6, “ Available
CacheNegotiatedD ocs options” . Since the content-negotiated documents may change over
time or because of the input from the requester, the default option is O f f .
T ab le 17.6 . Availab le C ach eN eg o t iat ed D o cs o p t io n s

366

⁠Chapt er 1 7 . Web Servers

O p t io n

D escrip t io n

On
Off

Enables caching the content-negotiated documents.
D isables caching the content-negotiated documents.

Examp le 17.26 . U sin g t h e C ach eN eg o t iat ed D o cs d irect ive
CacheNegotiatedD ocs On
C ach eR o o t
The C ach eR o o t directive allows you to specify the directory to store cache files in. It takes
the following form:
CacheRoot directory
The directory must be a full path to an existing directory in the local file system. The default
option is /var/cach e/mo d _p ro xy/.

Examp le 17.27. U sin g t h e C ach eR o o t d irect ive
CacheRoot /var/cache/mod_proxy
C u st o mLo g
The C u st o mLo g directive allows you to specify the log file name and the log file format. It
takes the following form:
CustomLog path format
The path refers to a log file, and must be relative to the directory that is specified by the
ServerR o o t directive (that is, /et c/h t t p d / by default). The format has to be either an
explicit format string, or a format name that was previously defined using the Lo g Fo rmat
directive.

Examp le 17.28. U sin g t h e C u st o mLo g d irect ive
CustomLog logs/access_log combined
D ef au lt Ico n
The D ef au lt Ico n directive allows you to specify an icon to be displayed for a file in servergenerated directory listings when no other icon is associated with it. It takes the following
form:
D efaultIcon path
The path refers to an existing icon file, and must be relative to the directory specified by the
D o cu men t R o o t directive (for example, /ico n s/u n kn o wn .p n g ).

Examp le 17.29 . U sin g t h e D ef au lt Ico n d irect ive

367

Deployment G uide

D efaultIcon /icons/unknown.png
D ef au lt T yp e
The D ef au lt T yp e directive allows you to specify a media type to be used in case the
proper MIME type cannot be determined by the server. It takes the following form:
D efaultType content-type
The content-type has to be a valid MIME type such as t ext /h t ml, imag e/p n g ,
ap p licat io n /p d f , etc.

Examp le 17.30. U sin g t h e D ef au lt T yp e d irect ive
D efaultType text/plain
D en y
The D en y directive allows you to specify which clients are denied access to a given
directory. It takes the following form:
D eny from client…
The client can be a domain name, an IP address (both full and partial), a network/netmask
pair, or all for all clients.

Examp le 17.31. U sin g t h e D en y d irect ive
D eny from 192.168.1.1
D irect o ryIn d ex
The D irect o ryIn d ex directive allows you to specify a document to be served to a client
when a directory is requested (that is, when the URL ends with the / character). It takes the
following form:
D irectoryIndex filename…
The filename is a name of the file to look for in the requested directory. By default, the server
looks for in d ex.h t ml, and in d ex.h t ml.var.

Examp le 17.32. U sin g t h e D irect o ryIn d ex d irect ive
D irectoryIndex index.html index.html.var
D o cu men t R o o t
The D o cu men t R o o t directive allows you to specify the main directory from which the
content is served. It takes the following form:

368

⁠Chapt er 1 7 . Web Servers

D ocumentRoot directory
The directory must be a full path to an existing directory in the local file system. The default
option is /var/www/h t ml/.

Examp le 17.33. U sin g t h e D o cu men t R o o t d irect ive
D ocumentRoot /var/www/html
Erro rD o cu men t
The Erro rD o cu men t directive allows you to specify a document or a message to be
displayed as a response to a particular error. It takes the following form:
ErrorD ocument error-code action
The error-code has to be a valid code such as 4 03 (Forbidden), 4 04 (Not Found), or 500
(Internal Server Error). The action can be either a URL (both local and external), or a
message string enclosed in double quotes (that is, ").

Examp le 17.34 . U sin g t h e Erro rD o cu men t d irect ive
ErrorD ocument 403 " Access D enied"
ErrorD ocument 404 /404-not_found.html
Erro rLo g
The Erro rLo g directive allows you to specify a file to which the server errors are logged. It
takes the following form:
ErrorLog path
The path refers to a log file, and can be either absolute, or relative to the directory that is
specified by the ServerR o o t directive (that is, /et c/h t t p d / by default). The default option is
lo g s/erro r_lo g

Examp le 17.35. U sin g t h e Erro rLo g d irect ive
ErrorLog logs/error_log
Ext en d ed St at u s
The Ext en d ed St at u s directive allows you to enable detailed server status information. It
takes the following form:
ExtendedStatus option
The option has to be a valid keyword as described in Table 17.7, “ Available ExtendedStatus
options” . The default option is O f f .

369

Deployment G uide

T ab le 17.7. Availab le Ext en d ed St at u s o p t io n s
O p t io n

D escrip t io n

On
Off

Enables generating the detailed server status.
D isables generating the detailed server status.

Examp le 17.36 . U sin g t h e Ext en d ed St at u s d irect ive
ExtendedStatus On
G ro u p
The G ro u p directive allows you to specify the group under which the h t t p d service will
run. It takes the following form:
Group group
The group has to be an existing UNIX group. The default option is ap ach e.
Note that G ro u p is no longer supported inside , and has been replaced by
the Su execU serG ro u p directive.

Examp le 17.37. U sin g t h e G ro u p d irect ive
Group apache
H ead erN ame
The H ead erN ame directive allows you to specify a file to be prepended to the beginning of
the server-generated directory listing. It takes the following form:
HeaderName filename
The filename is a name of the file to look for in the requested directory. By default, the server
looks for H EAD ER .h t ml.

Examp le 17.38. U sin g t h e H ead erN ame d irect ive
HeaderName HEAD ER.html
H o st n ameLo o ku p s
The H o st n ameLo o ku p s directive allows you to enable automatic resolving of IP
addresses. It takes the following form:
HostnameLookups option
The option has to be a valid keyword as described in Table 17.8, “ Available
HostnameLookups options” . To conserve resources on the server, the default option is O f f .
T ab le 17.8. Availab le H o st n ameLo o ku p s o p t io n s

370

⁠Chapt er 1 7 . Web Servers

O p t io n

D escrip t io n

On

Enables resolving the IP address for each connection so that the
host name can be logged. However, this also adds a significant
processing overhead.
Enables performing the double-reverse D NS lookup. In
comparison to the above option, this adds even more processing
overhead.
D isables resolving the IP address for each connection.

D o u b le

Off

Note that when the presence of host names is required in server log files, it is often possible
to use one of the many log analyzer tools that perform the D NS lookups more efficiently.

Examp le 17.39 . U sin g t h e H o st n ameLo o ku p s d irect ive
HostnameLookups Off
In clu d e
The In clu d e directive allows you to include other configuration files. It takes the following
form:
Include filename
The f ilen ame can be an absolute path, a path relative to the directory specified by the
ServerR o o t directive, or a wildcard expression. All configuration files from the
/et c/h t t p d /co n f .d / directory are loaded by default.

Examp le 17.4 0. U sin g t h e In clu d e d irect ive
Include conf.d/*.conf
In d exIg n o re
The In d exIg n o re directive allows you to specify a list of file names to be omitted from the
server-generated directory listings. It takes the following form:
IndexIgnore filename…
The filename option can be either a full file name, or a wildcard expression.

Examp le 17.4 1. U sin g t h e In d exIg n o re d irect ive
IndexIgnore .??* *~ *# HEAD ER* READ ME* RCS CVS *,v *,t
In d exO p t io n s
The In d exO p t io n s directive allows you to customize the behavior of server-generated
directory listings. It takes the following form:
IndexOptions option…

371

Deployment G uide

The option has to be a valid keyword as described in Table 17.9, “ Available directory listing
options” . The default options are C h arset = U T F- 8, Fan cyIn d exin g , H T MLT ab le,
N ameWid t h = *, and Versio n So rt .
T ab le 17.9 . Availab le d irect o ry list in g o p t io n s
O p t io n
C h arset =encoding

D escrip t io n

Specifies the character set of a generated web page. The
encoding has to be a valid character set such as U T F- 8 or
ISO - 8859 - 2.
T yp e=content-type
Specifies the media type of a generated web page. The
content-type has to be a valid MIME type such as t ext /h t ml
or t ext /p lain .
D escrip t io n Wid t h =value
Specifies the width of the description column. The value
can be either a number of characters, or an asterisk (that
is, *) to adjust the width automatically.
Fan cyIn d exin g
Enables advanced features such as different icons for
certain files or possibility to re-sort a directory listing by
clicking on a column header.
Fo ld erFirst
Enables listing directories first, always placing them above
files.
H T MLT ab le
Enables the use of HTML tables for directory listings.
Ico n sAreLin ks
Enables using the icons as links.
Ico n H eig h t =value
Specifies an icon height. The value is a number of pixels.
Ico n Wid t h =value
Specifies an icon width. The value is a number of pixels.
Ig n o reC ase
Enables sorting files and directories in a case-sensitive
manner.
Ig n o reC lien t
D isables accepting query variables from a client.
N ameWid t h =value
Specifies the width of the file name column. The value can
be either a number of characters, or an asterisk (that is, *)
to adjust the width automatically.
Scan H T MLT it les
Enables parsing the file for a description (that is, the t it le
element) in case it is not provided by the Ad d D escrip t io n
directive.
Sh o wFo rb id d en
Enables listing the files with otherwise restricted access.
Su p p ressC o lu mn So rt in D isables re-sorting a directory listing by clicking on a
g
column header.
Su p p ressD escrip t io n
D isables reserving a space for file descriptions.
Su p p ressH T MLPreamb le D isables the use of standard HTML preamble when a file
specified by the H ead erN ame directive is present.
Su p p ressIco n
D isables the use of icons in directory listings.
Su p p ressLast Mo d if ied
D isables displaying the date of the last modification field in
directory listings.
Su p p ressR u les
D isables the use of horizontal lines in directory listings.
Su p p ressSiz e
D isables displaying the file size field in directory listings.
T rackMo d if ied
Enables returning the Last - Mo d if ied and ET ag values in
the HTTP header.
Versio n So rt
Enables sorting files that contain a version number in the
expected manner.
XH T ML
Enables the use of XHTML 1.0 instead of the default HTML
3.2.

372

⁠Chapt er 1 7 . Web Servers

Examp le 17.4 2. U sin g t h e In d exO p t io n s d irect ive
IndexOptions FancyIndexing VersionSort NameWidth=* HTMLTable Charset=UTF-8
K eep Alive
The K eep Alive directive allows you to enable persistent connections. It takes the following
form:
KeepAlive option
The option has to be a valid keyword as described in Table 17.10, “ Available KeepAlive
options” . The default option is O f f .
T ab le 17.10. Availab le K eep Alive o p t io n s
O p t io n

D escrip t io n

On

Enables the persistent connections. In this case, the server will
accept more than one request per connection.
D isables the keep-alive connections.

Off

Note that when the persistent connections are enabled, on a busy server, the number of
child processes can increase rapidly and eventually reach the maximum limit, slowing
down the server significantly. To reduce the risk, it is recommended that you set
K eep AliveT imeo u t to a low number, and monitor the /var/lo g /h t t p d /lo g s/erro r_lo g log
file carefully.

Examp le 17.4 3. U sin g t h e K eep Alive d irect ive
KeepAlive Off
K eep AliveT imeo u t
The K eep AliveT imeo u t directive allows you to specify the amount of time to wait for
another request before closing the connection. It takes the following form:
KeepAliveTimeout time
The time is specified in seconds. The default option is 15.

Examp le 17.4 4 . U sin g t h e K eep AliveT imeo u t d irect ive
KeepAliveTimeout 15
Lan g u ag ePrio rit y
The Lan g u ag ePrio rit y directive allows you to customize the precedence of languages. It
takes the following form:
LanguagePriority language…

373

Deployment G uide

The language has to be a valid MIME language such as cs, en , or f r.
This directive is especially useful for web servers that serve content in multiple languages
based on the client's language settings.

Examp le 17.4 5. U sin g t h e Lan g u ag ePrio rit y d irect ive
LanguagePriority sk cs en
List en
The Listen directive allows you to specify IP addresses or ports to listen to. It takes the
following form:
Listen [ip-address:]port [protocol]
The ip-address is optional and unless supplied, the server will accept incoming requests on
a given port from all IP addresses. Since the protocol is determined automatically from the
port number, it can be usually omitted. The default option is to listen to port 80.
Note that if the server is configured to listen to a port under 1024, only superuser will be
able to start the h t t p d service.

Examp le 17.4 6 . U sin g t h e List en d irect ive
Listen 80
Lo ad Mo d u le
The Lo ad Mo d u le directive allows you to load a Dynamic Shared Object (D SO) module. It
takes the following form:
LoadModule name path
The name has to be a valid identifier of the required module. The path refers to an existing
module file, and must be relative to the directory in which the libraries are placed (that is,
/u sr/lib /h t t p d / on 32-bit and /u sr/lib 6 4 /h t t p d / on 64-bit systems by default).
See Section 17.1.6, “ Working with Modules” for more information on the Apache HTTP
Server's D SO support.

Examp le 17.4 7. U sin g t h e Lo ad Mo d u le d irect ive
LoadModule php5_module modules/libphp5.so
Lo g Fo rmat
The LogFormat directive allows you to specify a log file format. It takes the following form:
LogFormat format name

374

⁠Chapt er 1 7 . Web Servers

The format is a string consisting of options as described in Table 17.11, “ Common
LogFormat options” . The name can be used instead of the format string in the C u st o mLo g
directive.
T ab le 17.11. C o mmo n Lo g Fo rmat o p t io n s
O p t io n

D escrip t io n

%b
%h
%l

Represents the size of the response in bytes.
Represents the IP address or host name of a remote client.
Represents the remote log name if supplied. If not, a hyphen (that is, - ) is
used instead.
Represents the first line of the request string as it came from the browser or
client.
Represents the status code.
Represents the date and time of the request.
If the authentication is required, it represents the remote user. If not, a hyphen
(that is, - ) is used instead.
Represents the content of the HTTP header field. The common options include
% {R ef erer} (the URL of the web page that referred the client to the server)
and % {U ser- Ag en t } (the type of the web browser making the request).

%r
%s
%t
%u
% {field}

Examp le 17.4 8. U sin g t h e Lo g Fo rmat d irect ive
LogFormat " % h % l % u % t \" % r\" % >s % b" common
Lo g Level
The Lo g Level directive allows you to customize the verbosity level of the error log. It takes
the following form:
LogLevel option
The option has to be a valid keyword as described in Table 17.12, “ Available LogLevel
options” . The default option is warn .
T ab le 17.12. Availab le Lo g Level o p t io n s
O p t io n

D escrip t io n

emerg

Only the emergency situations when the server cannot perform its
work are logged.
All situations when an immediate action is required are logged.
All critical conditions are logged.
All error messages are logged.
All warning messages are logged.
Even normal, but still significant situations are logged.
Various informational messages are logged.
Various debugging messages are logged.

alert
crit
erro r
warn
n o t ice
in f o
d eb u g

Examp le 17.4 9 . U sin g t h e Lo g Level d irect ive
LogLevel warn

375

Deployment G uide

MaxK eep AliveR eq u est s
The MaxK eep AliveR eq u est s directive allows you to specify the maximum number of
requests for a persistent connection. It takes the following form:
MaxKeepAliveRequests number
A high number can improve the performance of the server. Note that using 0 allows unlimited
number of requests. The default option is 100.

Examp le 17.50. U sin g t h e MaxK eep AliveR eq u est s o p t io n
MaxKeepAliveRequests 100
N ameVirt u alH o st
The N ameVirt u alH o st directive allows you to specify the IP address and port number for
a name-based virtual host. It takes the following form:
NameVirtualHost ip-address[:port]
The ip-address can be either a full IP address, or an asterisk (that is, *) representing all
interfaces. Note that IPv6 addresses have to be enclosed in square brackets (that is, [ and
] ). The port is optional.
Name-based virtual hosting allows one Apache HTTP Server to serve different domains
without using multiple IP addresses.

Using secure HT T P connections
Name-based virtual hosts only work with non-secure HTTP connections. If using
virtual hosts with a secure server, use IP address-based virtual hosts instead.

Examp le 17.51. U sin g t h e N ameVirt u alH o st d irect ive
NameVirtualHost *:80
O p t io n s
The O p t io n s directive allows you to specify which server features are available in a
particular directory. It takes the following form:
Options option…
The option has to be a valid keyword as described in Table 17.13, “ Available server
features” .
T ab le 17.13. Availab le server f eat u res

376

⁠Chapt er 1 7 . Web Servers

O p t io n

D escrip t io n

ExecC G I
Fo llo wSymLin ks
In clu d es
In clu d esN O EXEC

Enables the execution of CGI scripts.
Enables following symbolic links in the directory.
Enables server-side includes.
Enables server-side includes, but does not allow the
execution of commands.
Enables server-generated directory listings.
Enables content-negotiated “ MultiViews” .
Enables following symbolic links in the directory
when both the link and the target file have the same
owner.
Enables all of the features above with the exception
of Mu lt iViews.
D isables all of the features above.

In d exes
Mu lt iViews
SymLin ksIf O wn erMat ch

All
None

Important
The SymLin ksIf O wn erMat ch option is not a security feature as it can be bypassed
by an attacker.

Examp le 17.52. U sin g t h e O p t io n s d irect ive
Options Indexes FollowSymLinks
O rd er
The O rd er directive allows you to specify the order in which the Allo w and D en y directives
are evaluated. It takes the following form:
Order option
The option has to be a valid keyword as described in Table 17.14, “ Available Order options” .
The default option is allo w,d en y.
T ab le 17.14 . Availab le O rd er o p t io n s
O p t io n

D escrip t io n

allo w,d en y
d en y,allo w

Allo w directives are evaluated first.
D en y directives are evaluated first.

Examp le 17.53. U sin g t h e O rd er d irect ive
Order allow,deny
Pid File
The Pid File directive allows you to specify a file to which the process ID (PID ) of the server
is stored. It takes the following form:

377

Deployment G uide

PidFile path
The path refers to a pid file, and can be either absolute, or relative to the directory that is
specified by the ServerR o o t directive (that is, /et c/h t t p d / by default). The default option is
ru n /h t t p d .p id .

Examp le 17.54 . U sin g t h e Pid File d irect ive
PidFile run/httpd.pid
Pro xyR eq u est s
The Pro xyR eq u est s directive allows you to enable forward proxy requests. It takes the
following form:
ProxyRequests option
The option has to be a valid keyword as described in Table 17.15, “ Available ProxyRequests
options” . The default option is O f f .
T ab le 17.15. Availab le Pro xyR eq u est s o p t io n s
O p t io n

D escrip t io n

On
Off

Enables forward proxy requests.
D isables forward proxy requests.

Examp le 17.55. U sin g t h e Pro xyR eq u est s d irect ive
ProxyRequests On
R ead meN ame
The R ead meN ame directive allows you to specify a file to be appended to the end of the
server-generated directory listing. It takes the following form:
ReadmeName filename
The filename is a name of the file to look for in the requested directory. By default, the server
looks for R EAD ME.h t ml.

Examp le 17.56 . U sin g t h e R ead meN ame d irect ive
ReadmeName READ ME.html
R ed irect
The R ed irect directive allows you to redirect a client to another URL. It takes the following
form:
Redirect [status] path url

378

⁠Chapt er 1 7 . Web Servers

The status is optional, and if provided, it has to be a valid keyword as described in
Table 17.16, “ Available status options” . The path refers to the old location, and must be
relative to the directory specified by the D o cu men t R o o t directive (for example, /d o cs).
The url refers to the current location of the content (for example,
h t t p ://d o cs.examp le.co m).
T ab le 17.16 . Availab le st at u s o p t io n s
St at u s

D escrip t io n

p erman en t

Indicates that the requested resource has been moved
permanently. The 301 (Moved Permanently) status code is
returned to a client.
Indicates that the requested resource has been moved only
temporarily. The 302 (Found) status code is returned to a client.
Indicates that the requested resource has been replaced. The 303
(See Other) status code is returned to a client.
Indicates that the requested resource has been removed
permanently. The 4 10 (Gone) status is returned to a client.

t emp
seeo t h er
gone

Note that for more advanced redirection techniques, you can use the mo d _rewrit e module
that is part of the Apache HTTP Server installation.

Examp le 17.57. U sin g t h e R ed irect d irect ive
Redirect permanent /docs http://docs.example.com
Scrip t Alias
The Scrip t Alias directive allows you to specify the location of CGI scripts. It takes the
following form:
ScriptAlias url-path real-path
The url-path must be relative to the directory specified by the D o cu men t R o o t directive (for
example, /cg i- b in /). The real-path is a full path to a file or directory in the local file system.
This directive is typically followed by the D irect o ry tag with additional permissions to
access the target directory. By default, the /cg i- b in / alias is created so that the scripts
located in the /var/www/cg i- b in / are accessible.
The Scrip t Alias directive is used for security reasons to prevent CGI scripts from being
viewed as ordinary text documents.

Examp le 17.58. U sin g t h e Scrip t Alias d irect ive
ScriptAlias /cgi-bin/ /var/www/cgi-bin/

AllowOverride None
Options None
Order allow,deny
Allow from all


379

Deployment G uide

ServerAd min
The ServerAd min directive allows you to specify the email address of the server
administrator to be displayed in server-generated web pages. It takes the following form:
ServerAdmin email
The default option is ro o t @ lo calh o st .
This directive is commonly set to web mast er@ hostname, where hostname is the address of
the server. Once set, alias web mast er to the person responsible for the web server in
/et c/aliases, and as superuser, run the n ewaliases command.

Examp le 17.59 . U sin g t h e ServerAd min d irect ive
ServerAdmin webmaster@penguin.example.com
ServerN ame
The ServerN ame directive allows you to specify the host name and the port number of a
web server. It takes the following form:
ServerName hostname[:port]
The hostname has to be a fully qualified domain name (FQD N) of the server. The port is
optional, but when supplied, it has to match the number specified by the List en directive.
When using this directive, make sure that the IP address and server name pair are included
in the /et c/h o st s file.

Examp le 17.6 0. U sin g t h e ServerN ame d irect ive
ServerName penguin.example.com:80
ServerR o o t
The ServerR o o t directive allows you to specify the directory in which the server operates. It
takes the following form:
ServerRoot directory
The directory must be a full path to an existing directory in the local file system. The default
option is /et c/h t t p d /.

Examp le 17.6 1. U sin g t h e ServerR o o t d irect ive
ServerRoot /etc/httpd
ServerSig n at u re
The ServerSig n at u re directive allows you to enable displaying information about the
server on server-generated documents. It takes the following form:

380

⁠Chapt er 1 7 . Web Servers

server on server-generated documents. It takes the following form:
ServerSignature option
The option has to be a valid keyword as described in Table 17.17, “ Available
ServerSignature options” . The default option is O n .
T ab le 17.17. Availab le ServerSig n at u re o p t io n s
O p t io n

D escrip t io n

On

Enables appending the server name and version to servergenerated pages.
D isables appending the server name and version to servergenerated pages.
Enables appending the server name, version, and the email
address of the system administrator as specified by the
ServerAd min directive to server-generated pages.

Off
EMail

Examp le 17.6 2. U sin g t h e ServerSig n at u re d irect ive
ServerSignature On
ServerT o ken s
The ServerT o ken s directive allows you to customize what information is included in the
Server response header. It takes the following form:
ServerTokens option
The option has to be a valid keyword as described in Table 17.18, “ Available ServerTokens
options” . The default option is O S.
T ab le 17.18. Availab le ServerT o ken s o p t io n s
O p t io n

D escrip t io n

Pro d
Majo r

Includes the product name only (that is, Ap ach e).
Includes the product name and the major version of the server (for
example, 2).
Includes the product name and the minor version of the server (for
example, 2.2).
Includes the product name and the minimal version of the server
(for example, 2.2.15).
Includes the product name, the minimal version of the server, and
the type of the operating system it is running on (for example,
R ed H at ).
Includes all the information above along with the list of loaded
modules.

Min o r
Min
OS

Fu ll

Note that for security reasons, it is recommended to reveal as little information about the
server as possible.

Examp le 17.6 3. U sin g t h e ServerT o ken s d irect ive

381

Deployment G uide

ServerTokens Prod
Su execU serG ro u p
The Su execU serG ro u p directive allows you to specify the user and group under which
the CGI scripts will be run. It takes the following form:
SuexecUserGroup user group
The user has to be an existing user, and the group must be a valid UNIX group.
For security reasons, the CGI scripts should not be run with root privileges. Note that in
, Su execU serG ro u p replaces the U ser and G ro u p directives.

Examp le 17.6 4 . U sin g t h e Su execU serG ro u p d irect ive
SuexecUserGroup apache apache
T imeo u t
The T imeo u t directive allows you to specify the amount of time to wait for an event before
closing a connection. It takes the following form:
Timeout time
The time is specified in seconds. The default option is 6 0.

Examp le 17.6 5. U sin g t h e T imeo u t d irect ive
Timeout 60
T yp esC o n f ig
The T yp esC o n f ig allows you to specify the location of the MIME types configuration file. It
takes the following form:
TypesConfig path
The path refers to an existing MIME types configuration file, and can be either absolute, or
relative to the directory that is specified by the ServerR o o t directive (that is, /et c/h t t p d / by
default). The default option is /et c/mime.t yp es.
Note that instead of editing /et c/mime.t yp es, the recommended way to add MIME type
mapping to the Apache HTTP Server is to use the Ad d T yp e directive.

Examp le 17.6 6 . U sin g t h e T yp esC o n f ig d irect ive
TypesConfig /etc/mime.types
U seC an o n icalN ame

382

⁠Chapt er 1 7 . Web Servers

The U seC an o n icalN ame allows you to specify the way the server refers to itself. It takes
the following form:
UseCanonicalName option
The option has to be a valid keyword as described in Table 17.19, “ Available
UseCanonicalName options” . The default option is O f f .
T ab le 17.19 . Availab le U seC an o n icalN ame o p t io n s
O p t io n

D escrip t io n

On

Enables the use of the name that is specified by the ServerN ame
directive.
D isables the use of the name that is specified by the ServerN ame
directive. The host name and port number provided by the
requesting client are used instead.
D isables the use of the name that is specified by the ServerN ame
directive. The host name determined by a reverse D NS lookup is
used instead.

Off

DNS

Examp le 17.6 7. U sin g t h e U seC an o n icalN ame d irect ive
UseCanonicalName Off
U ser
The U ser directive allows you to specify the user under which the h t t p d service will run. It
takes the following form:
User user
The user has to be an existing UNIX user. The default option is ap ach e.
For security reasons, the h t t p d service should not be run with root privileges. Note that
U ser is no longer supported inside , and has been replaced by the
Su execU serG ro u p directive.

Examp le 17.6 8. U sin g t h e U ser d irect ive
User apache
U serD ir
The U serD ir directive allows you to enable serving content from users' home directories. It
takes the following form:
UserD ir option
The option can be either a name of the directory to look for in user's home directory
(typically p u b lic_h t ml), or a valid keyword as described in Table 17.20, “ Available
UserD ir options” . The default option is d isab led .

383

Deployment G uide

T ab le 17.20. Availab le U serD ir o p t io n s
O p t io n

D escrip t io n

en ab led user…
d isab led [user…]

Enables serving content from home directories of given users.
D isables serving content from home directories, either for all users,
or, if a space separated list of users is supplied, for given users
only.

Set the correct permissions
In order for the web server to access the content, the permissions on relevant
directories and files must be set correctly. Make sure that all users are able to access
the home directories, and that they can access and read the content of the directory
specified by the U serD ir directive. For example:
~]# ch mo d a+ x /h o me/username/
~]# ch mo d a+ rx /h o me/username/p u b lic_h t ml/
All files in this directory must be set accordingly.

Examp le 17.6 9 . U sin g t h e U serD ir d irect ive
UserD ir public_html

1 7 .1 .5 .2 . Co m m o n ssl.co nf Dire ct ive s
The Secure Sockets Layer (SSL) directives allow you to customize the behavior of the Apache HTTP
Secure Server, and in most cases, they are configured appropriately during the installation. Be
careful when changing these settings, as incorrect configuration can lead to security vulnerabilities.
The following directive is commonly used in /et c/h t t p d /co n f .d /ssl.co n f :
Set En vIf
The Set En vIf directive allows you to set environment variables based on the headers of
incoming connections. It takes the following form:
SetEnvIf option pattern [!]variable[=value]…
The option can be either a HTTP header field, a previously defined environment variable
name, or a valid keyword as described in Table 17.21, “ Available SetEnvIf options” . The
pattern is a regular expression. The variable is an environment variable that is set when the
option matches the pattern. If the optional exclamation mark (that is, !) is present, the
variable is removed instead of being set.
T ab le 17.21. Availab le Set En vIf o p t io n s

384

O p t io n

D escrip t io n

R emo t e_H o st

Refers to the client's host name.

⁠Chapt er 1 7 . Web Servers

O p t io n

D escrip t io n

R emo t e_Ad d r
Server_Ad d r
R eq u est _Met h o d
R eq u est _Pro t o co
l
R eq u est _U R I

Refers
Refers
Refers
Refers

to
to
to
to

the client's IP address.
the server's IP address.
the request method (for example, G ET ).
the protocol name and version (for example, H T T P/1.1).

Refers to the requested resource.

The Set En vIf directive is used to disable HTTP keepalives, and to allow SSL to close the
connection without a closing notification from the client browser. This is necessary for
certain web browsers that do not reliably shut down the SSL connection.

Examp le 17.70. U sin g t h e Set En vIf d irect ive
SetEnvIf User-Agent " .*MSIE.*" \
nokeepalive ssl-unclean-shutdown \
downgrade-1.0 force-response-1.0

Note that for the /et c/h t t p d /co n f .d /ssl.co n f file to be present, the mod_ssl needs to be installed.
See Section 17.1.8, “ Setting Up an SSL Server” for more information on how to install and configure
an SSL server.

1 7 .1 .5 .3. Co m m o n Mult i-Pro ce ssing Mo dule Dire ct ive s
The Multi-Processing Module (MPM) directives allow you to customize the behavior of a particular MPM
specific server-pool. Since its characteristics differ depending on which MPM is used, the directives
are embedded in If Mo d u le. By default, the server-pool is defined for both the p ref o rk and wo rker
MPMs.
The following MPM directives are commonly used in /et c/h t t p d /co n f /h t t p d .co n f :
MaxC lien t s
The MaxC lien t s directive allows you to specify the maximum number of simultaneously
connected clients to process at one time. It takes the following form:
MaxClients number
A high number can improve the performance of the server, although it is not recommended
to exceed 256 when using the p ref o rk MPM.

Examp le 17.71. U sin g t h e MaxC lien t s d irect ive
MaxClients 256
MaxR eq u est sPerC h ild
The MaxR eq u est sPerC h ild directive allows you to specify the maximum number of
request a child process can serve before it dies. It takes the following form:
MaxRequestsPerChild number

385

Deployment G uide

Setting the number to 0 allows unlimited number of requests.
The MaxR eq u est sPerC h ild directive is used to prevent long-lived processes from
causing memory leaks.

Examp le 17.72. U sin g t h e MaxR eq u est sPerC h ild d irect ive
MaxRequestsPerChild 4000
MaxSp areServers
The MaxSp areServers directive allows you to specify the maximum number of spare child
processes. It takes the following form:
MaxSpareServers number
This directive is used by the p ref o rk MPM only.

Examp le 17.73. U sin g t h e MaxSp areServers d irect ive
MaxSpareServers 20
MaxSp areT h read s
The MaxSp areT h read s directive allows you to specify the maximum number of spare
server threads. It takes the following form:
MaxSpareThreads number
The number must be greater than or equal to the sum of Min Sp areT h read s and
T h read sPerC h ild . This directive is used by the wo rker MPM only.

Examp le 17.74 . U sin g t h e MaxSp areT h read s d irect ive
MaxSpareThreads 75
Min Sp areServers
The Min Sp areServers directive allows you to specify the minimum number of spare child
processes. It takes the following form:
MinSpareServers number
Note that a high number can create a heavy processing load on the server. This directive is
used by the p ref o rk MPM only.

Examp le 17.75. U sin g t h e Min Sp areServers d irect ive
MinSpareServers 5

386

⁠Chapt er 1 7 . Web Servers

Min Sp areT h read s
The Min Sp areT h read s directive allows you to specify the minimum number of spare
server threads. It takes the following form:
MinSpareThreads number
This directive is used by the wo rker MPM only.

Examp le 17.76 . U sin g t h e Min Sp areT h read s d irect ive
MinSpareThreads 75
St art Servers
The St art Servers directive allows you to specify the number of child processes to create
when the service is started. It takes the following form:
StartServers number
Since the child processes are dynamically created and terminated according to the current
traffic load, it is usually not necessary to change this value.

Examp le 17.77. U sin g t h e St art Servers d irect ive
StartServers 8
T h read sPerC h ild
The T h read sPerC h ild directive allows you to specify the number of threads a child
process can create. It takes the following form:
ThreadsPerChild number
This directive is used by the wo rker MPM only.

Examp le 17.78. U sin g t h e T h read sPerC h ild d irect ive
ThreadsPerChild 25

17.1.6. Working wit h Modules
Being a modular application, the h t t p d service is distributed along with a number of Dynamic Shared
Objects (D SOs), which can be dynamically loaded or unloaded at runtime as necessary. By default,
these modules are located in /u sr/lib /h t t p d /mo d u les/ on 32-bit and in
/u sr/lib 6 4 /h t t p d /mo d u les/ on 64-bit systems.

1 7 .1 .6 .1 . Lo ading a Mo dule

387

Deployment G uide

To load a particular D SO module, use the Lo ad Mo d u le directive as described in Section 17.1.5.1,
“ Common httpd.conf D irectives” . Note that modules provided by a separate package often have their
own configuration file in the /et c/h t t p d /co n f .d / directory.

Examp le 17.79 . Lo ad in g t h e mo d _ssl D SO
LoadModule ssl_module modules/mod_ssl.so

Once you are finished, restart the web server to reload the configuration. See Section 17.1.4.3,
“ Restarting the Service” for more information on how to restart the h t t p d service.

1 7 .1 .6 .2 . Writ ing a Mo dule
If you intend to create a new D SO module, make sure you have the httpd-devel package installed. To
do so, enter the following command as ro o t :
~]# yu m in st all h t t p d - d evel
This package contains the include files, the header files, and the APach e eXt en Sio n (ap xs) utility
required to compile a module.
Once written, you can build the module with the following command:
~]# ap xs - i - a - c module_name.c
If the build was successful, you should be able to load the module the same way as any other
module that is distributed with the Apache HTTP Server.

17.1.7. Set t ing Up Virt ual Host s
The Apache HTTP Server's built in virtual hosting allows the server to provide different information
based on which IP address, host name, or port is being requested.
To create a name-based virtual host, find the virtual host container provided in
/et c/h t t p d /co n f /h t t p d .co n f as an example, remove the hash sign (that is, #) from the beginning of
each line, and customize the options according to your requirements as shown in Example 17.80,
“ Example virtual host configuration” .

Examp le 17.80. Examp le virt u al h o st co n f ig u rat io n
NameVirtualHost *:80

ServerAdmin webmaster@penguin.example.com
D ocumentRoot /www/docs/penguin.example.com
ServerName penguin.example.com
ServerAlias www.penguin.example.com
ErrorLog logs/penguin.example.com-error_log
CustomLog logs/penguin.example.com-access_log common


388

⁠Chapt er 1 7 . Web Servers

Note that ServerN ame must be a valid D NS name assigned to the machine. The 
container is highly customizable, and accepts most of the directives available within the main server
configuration. D irectives that are not supported within this container include U ser and G ro u p ,
which were replaced by Su execU serG ro u p .

Changing the port number
If you configure a virtual host to listen on a non-default port, make sure you update the List en
directive in the global settings section of the /et c/h t t p d /co n f /h t t p d .co n f file accordingly.
To activate a newly created virtual host, the web server has to be restarted first. See Section 17.1.4.3,
“ Restarting the Service” for more information on how to restart the h t t p d service.

17.1.8. Set t ing Up an SSL Server
Secure Sockets Layer (SSL) is a cryptographic protocol that allows a server and a client to
communicate securely. Along with its extended and improved version called Transport Layer Security
(TLS), it ensures both privacy and data integrity. The Apache HTTP Server in combination with
mo d _ssl, a module that uses the OpenSSL toolkit to provide the SSL/TLS support, is commonly
referred to as the SSL server. Red Hat Enterprise Linux also supports the use of Mozilla NSS as the
TLS implementation. Support for Mozilla NSS is provided by the mo d _n ss module.
Unlike an HTTP connection that can be read and possibly modified by anybody who is able to
intercept it, the use of SSL/TLS over HTTP, referred to as HTTPS, prevents any inspection or
modification of the transmitted content. This section provides basic information on how to enable this
module in the Apache HTTP Server configuration, and guides you through the process of generating
private keys and self-signed certificates.

1 7 .1 .8 .1 . An Ove rvie w o f Ce rt ificat e s and Se curit y
Secure communication is based on the use of keys. In conventional or symmetric cryptography, both
ends of the transaction have the same key they can use to decode each other's transmissions. On
the other hand, in public or asymmetric cryptography, two keys co-exist: a private key that is kept a
secret, and a public key that is usually shared with the public. While the data encoded with the public
key can only be decoded with the private key, data encoded with the private key can in turn only be
decoded with the public key.
To provide secure communications using SSL, an SSL server must use a digital certificate signed by
a Certificate Authority (CA). The certificate lists various attributes of the server (that is, the server host
name, the name of the company, its location, etc.), and the signature produced using the CA's private
key. This signature ensures that a particular certificate authority has signed the certificate, and that
the certificate has not been modified in any way.
When a web browser establishes a new SSL connection, it checks the certificate provided by the web
server. If the certificate does not have a signature from a trusted CA, or if the host name listed in the
certificate does not match the host name used to establish the connection, it refuses to communicate
with the server and usually presents a user with an appropriate error message.
By default, most web browsers are configured to trust a set of widely used certificate authorities.
Because of this, an appropriate CA should be chosen when setting up a secure server, so that target
users can trust the connection, otherwise they will be presented with an error message, and will have
to accept the certificate manually. Since encouraging users to override certificate errors can allow an
attacker to intercept the connection, you should use a trusted CA whenever possible. For more
information on this, see Table 17.22, “ Information about CA lists used by common web browsers” .

389

Deployment G uide

T ab le 17.22. In f o rmat io n ab o u t C A list s u sed b y co mmo n web b ro wsers
Web B ro wser

Lin k

Mo z illa Firef o x
O p era
In t ern et Exp lo rer
C h ro miu m

Mozilla root CA list.
Information on root certificates used by Opera.
Information on root certificates used by Microsoft Windows.
Information on root certificates used by the Chromium project.

When setting up an SSL server, you need to generate a certificate request and a private key, and then
send the certificate request, proof of the company's identity, and payment to a certificate authority.
Once the CA verifies the certificate request and your identity, it will send you a signed certificate you
can use with your server. Alternatively, you can create a self-signed certificate that does not contain a
CA signature, and thus should be used for testing purposes only.

17.1.9. Enabling t he mod_ssl Module
If you intend to set up an SSL or HTTPS server using mo d _ssl, you can n o t have another
application or module, such as mo d _n ss configured to use the same port. Port 4 4 3 is the default
port for HTTPS.
To set up an SSL server using the mo d _ssl module and the OpenSSL toolkit, install the mod_ssl and
openssl packages. Enter the following command as ro o t :
~]# yu m in st all mo d _ssl o p en ssl
This will create the mo d _ssl configuration file at /et c/h t t p d /co n f .d /ssl.co n f , which is included in
the main Apache HTTP Server configuration file by default. For the module to be loaded, restart the
h t t p d service as described in Section 17.1.4.3, “ Restarting the Service” .

Important
D ue to the vulnerability described in POODLE: SSLv3 vulnerability (CVE-2014-3566), Red Hat
recommends disabling SSL and using only T LSv1.1 or T LSv1.2. Backwards compatibility
can be achieved using T LSv1.0. Many products Red Hat supports have the ability to use
SSLv2 or SSLv3 protocols, or enable them by default. However, the use of SSLv2 or SSLv3 is
now strongly recommended against.

1 7 .1 .9 .1 . Enabling and Disabling SSL and T LS in m o d_ssl
To disable and enable specific versions of the SSL and TLS protocol, either do it globally by adding
the SSLPro t o co l directive in the “ ## SSL Global Context” section of the configuration file and
removing it everywhere else, or edit the default entry under “ # SSL Protocol support” in all
“ VirtualHost” sections. If you do not specify it in the per-domain VirtualHost section then it will inherit
the settings from the global section. To make sure that a protocol version is being disabled the
administrator should either o n ly specify SSLPro t o co l in the “ SSL Global Context” section, or
specify it in all per-domain VirtualHost sections.
Pro ced u re 17.1. D isab le SSLv2 an d SSLv3
To disable SSL version 2 and SSL version 3, which implies enabling everything except SSL version
2 and SSL version 3, in all VirtualHost sections, proceed as follows:

390

⁠Chapt er 1 7 . Web Servers

1. As ro o t , open the /et c/h t t p d /co n f .d /ssl.co n f file and search for all instances of the
SSLPro t o co l directive. By default, the configuration file contains one section that looks as
follows:
~]# vi /et c/h t t p d /co n f .d /ssl.co n f
# SSL Protocol support:
# List the enable protocol levels with which clients will be able to
# connect. D isable SSLv2 access by default:
SSLProtocol all -SSLv2
This section is within the VirtualHost section.
2. Edit the SSLPro t o co l line as follows:
# SSL Protocol support:
# List the enable protocol levels with which clients will be able to
# connect. D isable SSLv2 access by default:
SSLProtocol all -SSLv2 -SSLv3
Repeat this action for all VirtualHost sections. Save and close the file.
3. Verify that all occurrences of the SSLPro t o co l directive have been changed as follows:
~]# g rep SSLPro t o co l /et c/h t t p d /co n f .d /ssl.co n f
SSLPro t o co l all -SSLv2 -SSLv3
This step is particularly important if you have more than the one default VirtualHost section.
4. Restart the Apache daemon as follows:
~]# service h t t p d rest art
Note that any sessions will be interrupted.
Pro ced u re 17.2. D isab le All SSL an d T LS Pro t o co ls Excep t T LS 1 an d U p
To disable all SSL and TLS protocol versions except TLS version 1 and higher, proceed as follows:
1. As ro o t , open the /et c/h t t p d /co n f .d /ssl.co n f file and search for all instances of
SSLPro t o co l directive. By default the file contains one section that looks as follows:
~]# vi /et c/h t t p d /co n f .d /ssl.co n f
# SSL Protocol support:
# List the enable protocol levels with which clients will be able to
# connect. D isable SSLv2 access by default:
SSLProtocol all -SSLv2
2. Edit the SSLPro t o co l line as follows:
# SSL Protocol support:
# List the enable protocol levels with which clients will be able to
# connect. D isable SSLv2 access by default:
SSLProtocol -all +TLSv1 +TLSv1.1 +TLSv1.2
Save and close the file.

391

Deployment G uide

3. Verify the change as follows:
~]# g rep SSLPro t o co l /et c/h t t p d /co n f .d /ssl.co n f
SSLPro t o co l -all +TLSv1 +TLSv1.1 +TLSv1.2
4. Restart the Apache daemon as follows:
~]# service h t t p d rest art
Note that any sessions will be interrupted.
Pro ced u re 17.3. T est in g t h e St at u s o f SSL an d T LS Pro t o co ls
To check which versions of SSL and TLS are enabled or disabled, make use of the o p en ssl
s_clien t - co n n ect command. The command has the following form:
openssl s_client -connect hostname:port -protocol
Where port is the port to test and protocol is the protocol version to test for. To test the SSL server
running locally, use lo calh o st as the host name. For example, to test the default port for secure
HTTPS connections, port 4 4 3 to see if SSLv3 is enabled, issue a command as follows:
1.

~]# o p en ssl s_clien t - co n n ect lo calh o st :4 4 3 - ssl3
CONNECTED (00000003)
139809943877536:error:14094410:SSL routines:SSL3_READ _BYTES:sslv3 alert
h an d sh ake f ailu re:s3_pkt.c:1257:SSL alert number 40
139809943877536:error:1409E0E5:SSL routines:SSL3_WRITE_BYTES:ssl h an d sh ake
f ailu re:s3_pkt.c:596:
output omitted
New, (NONE), C ip h er is ( N O N E)
Secure Renegotiation IS NOT supported
Compression: NONE
Expansion: NONE
SSL-Session:
Protocol : SSLv3
output truncated
The above output indicates that the handshake failed and therefore no cipher was
negotiated.

2.

392

~]$ o p en ssl s_clien t - co n n ect lo calh o st :4 4 3 - t ls1_2
CONNECTED (00000003)
depth=0 C = --, ST = SomeState, L = SomeCity, O = SomeOrganization, OU =
SomeOrganizationalUnit, CN = localhost.localdomain, emailAddress =
root@localhost.localdomain
output omitted
New, TLSv1/SSLv3, C ip h er is EC D H E- R SA- AES256 - G C M- SH A384
Server public key is 2048 bit
Secure Renegotiation IS supported
Compression: NONE
Expansion: NONE
SSL-Session:
Protocol : TLSv1.2
output truncated

⁠Chapt er 1 7 . Web Servers

The above output indicates that no failure of the handshake occurred and a set of ciphers
was negotiated.
The o p en ssl s_clien t command options are documented in the s_clien t ( 1) manual page.
For more information on the SSLv3 vulnerability and how to test for it, see the Red Hat
Knowledgebase article POODLE: SSLv3 vulnerability (CVE-2014-3566).

17.1.10. Enabling t he mod_nss Module
If you intend to set up an HTTPS server using mo d _n ss, the HTTPS server can n o t simultaneously
use mo d _ssl with its default settings as mo d _ssl will use port 4 4 3 by default, however this is the
default HTTPS port. If is recommend to remove the package if it is not required.
To remove mod_ssl, enter the following command as ro o t :
~]# yu m remo ve mo d _ssl

Note
If mo d _ssl is required for other purposes, modify the /et c/h t t p d /co n f .d /ssl.co n f file to use
a port other than 4 4 3 to prevent mo d _ssl conflicting with mo d _n ss when its port to listen on
is changed to 4 4 3.
For a specific VirtualHost where HTTPS is required, mo d _n ss and mo d _ssl can only co-exist
at the same time if they use unique ports. For this reason mo d _n ss by default uses 84 4 3, but
the default port for HTTPS is port 4 4 3. The port is specified by the List en directive as well as
in the VirtualHost name or address.

Everything in NSS is associated with a “ token” . The software token exists in the NSS database but
you can also have a physical token containing certificates. With OpenSSL, discrete certificates and
private keys are held in PEM files. With NSS, these are stored in a database. Each certificate and key
is associated with a token and each token can have a password protecting it. This password is
optional, but if a password is used then the Apache HTTP server needs a copy of it in order to open
the database without user intervention at system start.
Pro ced u re 17.4 . C o n f ig u rin g mo d _n ss
1. Install mod_nss as ro o t :
~]# yu m in st all mo d _n ss
This will create the mo d _n ss configuration file at /et c/h t t p d /co n f .d /n ss.co n f . The
/et c/h t t p d /co n f .d / directory is included in the main Apache HTTP Server configuration file
by default. For the module to be loaded, restart the h t t p d service as described in
Section 17.1.4.3, “ Restarting the Service” .
2. As ro o t , open the /et c/h t t p d /co n f .d /n ss.co n f file and search for all instances of the
List en directive.
Edit the List en 84 4 3 line as follows:
Listen 443

393

Deployment G uide

Port 4 4 3 is the default port for H T T PS.
3. Edit the default Virt u alH o st _d ef au lt _:84 4 3 line as follows:
VirtualHost _default_:443
Edit any other non-default virtual host sections if they exist. Save and close the file.
4. Mozilla NSS stores certificates in a server certificate database indicated by the
N SSC ert if icat eD at ab ase directive in the /et c/h t t p d /co n f .d /n ss.co n f file. By default the
path is set to /et c/h t t p d /alias, the NSS database created during installation.
To view the default NSS database, issue a command as follows:
~]# cert u t il - L - d /et c/h t t p d /alias
Certificate Nickname

cacert
Server-Cert
alpha

Trust Attributes
SSL,S/MIME,JAR/XPI
CTu,Cu,Cu
u,u,u
u,pu,u

In the above command output, Server- C ert is the default N SSN ickn ame. The - L option lists
all the certificates, or displays information about a named certificate, in a certificate database.
The - d option specifies the database directory containing the certificate and key database
files. See the cert u t il( 1) man page for more command line options.
5. To configure mod_nss to use another database, edit the N SSC ert if icat eD at ab ase line in
the /et c/h t t p d /co n f .d /n ss.co n f file. The default file has the following lines within the
VirtualHost section.
# Server Certificate D atabase:
# The NSS security database directory that holds the certificates and
# keys. The database consists of 3 files: cert8.db, key3.db and secmod.db.
# Provide the directory that these files exist.
NSSCertificateD atabase /etc/httpd/alias
In the above command output, alias is the default NSS database directory,
/et c/h t t p d /alias/.
6. To apply a password to the default NSS certificate database, use the following command as
ro o t :
~]# cert u t il - W - d /et c/h t t p d /alias
Enter Password or Pin for " NSS Certificate D B" :
Enter a password which will be used to encrypt your keys.
The password should be at least 8 characters long,
and should contain at least one non-alphabetic character.
Enter new password:
Re-enter password:
Password changed successfully.
7. Before deploying the HTTPS server, create a new certificate database using a certificate
signed by a certificate authority (CA).

394

⁠Chapt er 1 7 . Web Servers

Examp le 17.81. Ad d in g a C ert if icat e t o t h e Mo z illa N SS d at ab ase
The cert u t il command is used to add a CA certificate to the NSS database files:
cert u t il - d /et c/h t t p d /nss-db-directory/ - A - n " CA_certificate" - t C T ,, - a - i
cert if icat e.p em
The above command adds a CA certificate stored in a PEM-formatted file named
certificate.pem. The - d option specifies the NSS database directory containing the
certificate and key database files, the - n option sets a name for the certificate, - t C T ,,
means that the certificate is trusted to be used in TLS clients and servers. The - A option
adds an existing certificate to a certificate database. If the database does not exist it will be
created. The - a option allows the use of ASCII format for input or output, and the - i option
passes the cert if icat e.p em input file to the command.
See the cert u t il( 1) man page for more command line options.
8. The NSS database should be password protected to safeguard the private key.

Examp le 17.82. Set t in g _a_Passwo rd _f o r_a_Mo z illa_N SS_d at ab ase
The cert u t il tool can be used set a password for an NSS database as follows:
certutil -W -d /etc/httpd/nss-db-directory/
For example, for the default database, issue a command as ro o t as follows:
~]# cert u t il - W - d /et c/h t t p d /alias
Enter Password or Pin for " NSS Certificate D B" :
Enter a password which will be used to encrypt your keys.
The password should be at least 8 characters long,
and should contain at least one non-alphabetic character.
Enter new password:
Re-enter password:
Password changed successfully.

9. Configure mo d _n ss to use the NSS internal software token by changing the line with the
N SSPassPh raseD ialo g directive as follows:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
NSSPassPhraseD ialog file:/etc/httpd/password.conf
This is to avoid manual password entry on system start. The software token exists in the NSS
database but you can also have a physical token containing your certificates.
10. If the SSL Server Certificate contained in the NSS database is an RSA certificate, make certain
that the N SSN ickn ame parameter is uncommented and matches the nickname displayed in
step 4 above:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
NSSNickname Server-Cert

395

Deployment G uide

If the SSL Server Certificate contained in the NSS database is an ECC certificate, make certain
that the N SSEC C N ickn ame parameter is uncommented and matches the nickname
displayed in step 4 above:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
NSSECCNickname Server-Cert
Make certain that the N SSC ert if icat eD at ab ase parameter is uncommented and points to
the NSS database directory displayed in step 4 or configured in step 5 above:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
NSSCertificateD atabase /etc/httpd/alias
Replace /et c/h t t p d /alias with the path to the certificate database to be used.
11. Create the /et c/h t t p d /p asswo rd .co n f file as ro o t :
~]# vi /et c/h t t p d /p asswo rd .co n f
Add a line with the following form:
internal:password
Replacing password with the password that was applied to the NSS security databases in
step 6 above.
12. Apply the appropriate ownership and permissions to the /et c/h t t p d /p asswo rd .co n f file:
~]# ch g rp ap ach e /et c/h t t p d /p asswo rd .co n f
~]# chmod 640 /etc/httpd/password.conf
~]# ls -l /etc/httpd/password.conf
-rw-r-----. 1 root apache 10 D ec 4 17:13 /etc/httpd/password.conf
13. To configure mo d _n ss to use the NSS the software token in /et c/h t t p d /p asswo rd .co n f ,
edit /et c/h t t p d /co n f .d /n ss.co n f as follows:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
14. Restart the Apache server for the changes to take effect as described in Section 17.1.4.3,
“ Restarting the Service”

Important
D ue to the vulnerability described in POODLE: SSLv3 vulnerability (CVE-2014-3566), Red Hat
recommends disabling SSL and using only T LSv1.1 or T LSv1.2. Backwards compatibility
can be achieved using T LSv1.0. Many products Red Hat supports have the ability to use
SSLv2 or SSLv3 protocols, or enable them by default. However, the use of SSLv2 or SSLv3 is
now strongly recommended against.

1 7 .1 .1 0 .1 . Enabling and Disabling SSL and T LS in m o d_nss

396

⁠Chapt er 1 7 . Web Servers

To disable and enable specific versions of the SSL and TLS protocol, either do it globally by adding
the N SSPro t o co l directive in the “ ## SSL Global Context” section of the configuration file and
removing it everywhere else, or edit the default entry under “ # SSL Protocol” in all “ VirtualHost”
sections. If you do not specify it in the per-domain VirtualHost section then it will inherit the settings
from the global section. To make sure that a protocol version is being disabled the administrator
should either o n ly specify N SSPro t o co l in the “ SSL Global Context” section, or specify it in all perdomain VirtualHost sections.
Pro ced u re 17.5. D isab le All SSL an d T LS Pro t o co ls Excep t T LS 1 an d U p in mo d _n ss
To disable all SSL and TLS protocol versions except TLS version 1 and higher, proceed as follows:
1. As ro o t , open the /et c/h t t p d /co n f .d /n ss.co n f file and search for all instances of the
N SSPro t o co l directive. By default, the configuration file contains one section that looks as
follows:
~]# vi /et c/h t t p d /co n f .d /n ss.co n f
# SSL Protocol:
output omitted
# Since all protocol ranges are completely inclusive, and no protocol in the
# middle of a range may be excluded, the entry " NSSProtocol SSLv3,TLSv1.1"
# is identical to the entry " NSSProtocol SSLv3,TLSv1.0,TLSv1.1" .
NSSProtocol SSLv3,TLSv1.0,TLSv1.1
This section is within the VirtualHost section.
2. Edit the N SSPro t o co l line as follows:
# SSL Protocol:
NSSProtocol TLSv1.0,TLSv1.1
Repeat this action for all VirtualHost sections.
3. Edit the List en 84 4 3 line as follows:
Listen 443
4. Edit the default Virt u alH o st _d ef au lt _:84 4 3 line as follows:
VirtualHost _default_:443
Edit any other non-default virtual host sections if they exist. Save and close the file.
5. Verify that all occurrences of the N SSPro t o co l directive have been changed as follows:
~]# g rep N SSPro t o co l /et c/h t t p d /co n f .d /n ss.co n f
# middle of a range may be excluded, the entry " N SSPro t o co l SSLv3,TLSv1.1"
# is identical to the entry " N SSPro t o co l SSLv3,TLSv1.0,TLSv1.1" .
N SSPro t o co l TLSv1.0,TLSv1.1
This step is particularly important if you have more than one VirtualHost section.
6. Restart the Apache daemon as follows:
~]# service h t t p d rest art

397

Deployment G uide

Note that any sessions will be interrupted.
Pro ced u re 17.6 . T est in g t h e St at u s o f SSL an d T LS Pro t o co ls in mo d _n ss
To check which versions of SSL and TLS are enabled or disabled in mo d _n ss, make use of the
o p en ssl s_clien t - co n n ect command. Install the openssl package as ro o t :
~]# yu m in st all o p en ssl
The o p en ssl s_clien t - co n n ect command has the following form:
openssl s_client -connect hostname:port -protocol
Where port is the port to test and protocol is the protocol version to test for. To test the SSL server
running locally, use lo calh o st as the host name. For example, to test the default port for secure
HTTPS connections, port 4 4 3 to see if SSLv3 is enabled, issue a command as follows:
1.

~]# o p en ssl s_clien t - co n n ect lo calh o st :4 4 3 - ssl3
CONNECTED (00000003)
3077773036:error:1408F10B:SSL routines:SSL3_GET_RECORD :wrong version
number:s3_pkt.c:337:
output omitted
New, (NONE), C ip h er is ( N O N E)
Secure Renegotiation IS NOT supported
Compression: NONE
Expansion: NONE
SSL-Session:
Protocol : SSLv3
output truncated
The above output indicates that the handshake failed and therefore no cipher was
negotiated.

2.

~]$ o p en ssl s_clien t - co n n ect lo calh o st :4 4 3 - t ls1
CONNECTED (00000003)
depth=1 C = US, O = example.com, CN = Certificate Shack
output omitted
New, TLSv1/SSLv3, C ip h er is AES128- SH A
Server public key is 1024 bit
Secure Renegotiation IS supported
Compression: NONE
Expansion: NONE
SSL-Session:
Protocol : TLSv1
output truncated
The above output indicates that no failure of the handshake occurred and a set of ciphers
was negotiated.

The o p en ssl s_clien t command options are documented in the s_clien t ( 1) manual page.
For more information on the SSLv3 vulnerability and how to test for it, see the Red Hat
Knowledgebase article POODLE: SSLv3 vulnerability (CVE-2014-3566).

17.1.11. Using an Exist ing Key and Cert ificat e

398

⁠Chapt er 1 7 . Web Servers

If you have a previously created key and certificate, you can configure the SSL server to use these
files instead of generating new ones. There are only two situations where this is not possible:
1. You are changing the IP address or domain name.
Certificates are issued for a particular IP address and domain name pair. If one of these
values changes, the certificate becomes invalid.
2. You have a certificate from VeriSign, and you are changing the server software.
VeriSign, a widely used certificate authority, issues certificates for a particular software
product, IP address, and domain name. Changing the software product renders the certificate
invalid.
In either of the above cases, you will need to obtain a new certificate. For more information on this
topic, see Section 17.1.12, “ Generating a New Key and Certificate” .
If you want to use an existing key and certificate, move the relevant files to the /et c/p ki/t ls/p rivat e/
and /et c/p ki/t ls/cert s/ directories respectively. You can do so by issuing the following commands
as ro o t :
~]# mv key_file.key /et c/p ki/t ls/p rivat e/hostname.key
~]# mv certificate.crt /et c/p ki/t ls/cert s/hostname.crt
Then add the following lines to the /et c/h t t p d /co n f .d /ssl.co n f configuration file:
SSLCertificateFile /etc/pki/tls/certs/hostname.crt
SSLCertificateKeyFile /etc/pki/tls/private/hostname.key
To load the updated configuration, restart the h t t p d service as described in Section 17.1.4.3,
“ Restarting the Service” .

Examp le 17.83. U sin g a key an d cert if icat e f ro m t h e R ed H at Secu re Web Server
~]# mv /et c/h t t p d /co n f /h t t p sd .key /et c/p ki/t ls/p rivat e/p en g u in .examp le.co m.key
~]# mv /et c/h t t p d /co n f /h t t p sd .crt /et c/p ki/t ls/cert s/p en g u in .examp le.co m.crt

17.1.12. Generat ing a New Key and Cert ificat e
In order to generate a new key and certificate pair, the crypto-utils package must be installed on the
system. In Red Hat Enterprise Linux 6, the mod_ssl package is required by the g en key utility. To
install these, enter the following command as ro o t :
~]# yu m in st all cryp t o - u t ils mo d _ssl
This package provides a set of tools to generate and manage SSL certificates and private keys, and
includes g en key, the Red Hat Keypair Generation utility that will guide you through the key
generation process.

399

Deployment G uide

Replacing an existing certificate
If the server already has a valid certificate and you are replacing it with a new one, specify a
different serial number. This ensures that client browsers are notified of this change, update to
this new certificate as expected, and do not fail to access the page. To create a new certificate
with a custom serial number, use the following command instead of g en key:
~]# o p en ssl req - x509 - n ew - set _serial number - key hostname.key - o u t
hostname.crt

Remove a previously created key
If there already is a key file for a particular host name in your system, g en key will refuse to
start. In this case, remove the existing file using the following command as ro o t :
~]# rm /et c/p ki/t ls/p rivat e/hostname.key

To run the utility enter the g en key command as ro o t , followed by the appropriate host name (for
example, p en g u in .examp le.co m):
~]# g en key hostname
To complete the key and certificate creation, take the following steps:
1. Review the target locations in which the key and certificate will be stored.

Fig u re 17.1. R u n n in g t h e g en key u t ilit y

4 00

⁠Chapt er 1 7 . Web Servers

Use the T ab key to select the N ext button, and press En t er to proceed to the next screen.
2. Using the u p and d o wn arrow keys, select a suitable key size. Note that while a larger key
increases the security, it also increases the response time of your server. The NIST
recommends using 204 8 b it s. See NIST Special Publication 800-131A.

Fig u re 17.2. Select in g t h e key siz e
Once finished, use the T ab key to select the N ext button, and press En t er to initiate the
random bits generation process. D epending on the selected key size, this may take some
time.
3. D ecide whether you want to send a certificate request to a certificate authority.

4 01

Deployment G uide

Fig u re 17.3. G en erat in g a cert if icat e req u est
Use the T ab key to select Yes to compose a certificate request, or N o to generate a selfsigned certificate. Then press En t er to confirm your choice.
4. Using the Sp aceb ar key, enable ([ *] ) or disable ([ ] ) the encryption of the private key.

Fig u re 17.4 . En cryp t in g t h e p rivat e key
Use the T ab key to select the N ext button, and press En t er to proceed to the next screen.

4 02

⁠Chapt er 1 7 . Web Servers

5. If you have enabled the private key encryption, enter an adequate passphrase. Note that for
security reasons, it is not displayed as you type, and it must be at least five characters long.

Fig u re 17.5. En t erin g a p assp h rase
Use the T ab key to select the N ext button, and press En t er to proceed to the next screen.

Do not forget the passphrase
Entering the correct passphrase is required in order for the server to start. If you lose it,
you will need to generate a new key and certificate.
6. Customize the certificate details.

4 03

Deployment G uide

Fig u re 17.6 . Sp ecif yin g cert if icat e in f o rmat io n
Use the T ab key to select the N ext button, and press En t er to finish the key generation.
7. If you have previously enabled the certificate request generation, you will be prompted to
send it to a certificate authority.

Fig u re 17.7. In st ru ct io n s o n h o w t o sen d a cert if icat e req u est
Press En t er to return to a shell prompt.

4 04

⁠Chapt er 1 7 . Web Servers

Once generated, add the key and certificate locations to the /et c/h t t p d /co n f .d /ssl.co n f
configuration file:
SSLCertificateFile /etc/pki/tls/certs/hostname.crt
SSLCertificateKeyFile /etc/pki/tls/private/hostname.key
Finally, restart the h t t p d service as described in Section 17.1.4.3, “ Restarting the Service” , so that
the updated configuration is loaded.

17.1.13. Configure t he Firewall for HT T P and HT T PS Using t he Command Line
Red Hat Enterprise Linux does not allow H T T P and H T T PS traffic by default. To enable the system to
act as a web server, enable ports and protocols as required. The default port for H T T P is 80 and the
default port for H T T PS is 4 4 3. In both cases the T C P should be allowed to pass through the
firewall.
To enable port 80 for H T T P using the command line, issue the following command as ro o t :
~]# lo kkit - - p o rt = 80:t cp - - u p d at e
Note that this will restart the firewall as long as it has not been disabled with the - - d isab led option.
Active connections will be terminated and time out on the initiating machine. Use the lo kkit - - h elp
command to view the built in help.
To enable port 4 4 3 for H T T PS using the command line, issue the following command as ro o t :
~]# lo kkit - - p o rt = 4 4 3:t cp - - u p d at e
Note that this will restart the firewall as long as it has not been disabled with the - - d isab led option.
Active connections will be terminated and time out on the initiating machine. See the /et c/services
file for list of services and their associated ports.
When preparing a configuration file for multiple installations using administration tools, it is useful to
edit the firewall configuration file directly. Note that any mistakes in the configuration file could have
unexpected consequences, cause an error, and prevent the firewall settings from being applied.
Therefore, check the /et c/sysco n f ig /syst em- co n f ig - f irewall file thoroughly after editing. To apply
the settings in /et c/sysco n f ig /syst em- co n f ig - f irewall, issue the following command as ro o t :
~]# lo kkit - - u p d at e
For example, to enable H T T PS to pass through the firewall, by editing the configuration file, become
the ro o t user and add the following line to /et c/sysco n f ig /syst em- co n f ig - f irewall:
--port=443:tcp
Note that these changes will not take effect even if the firewall is reloaded or the system rebooted. To
apply the changes in /et c/sysco n f ig /syst em- co n f ig - f irewall, issue the following command as
ro o t :
~]# lo kkit - - u p d at e

1 7 .1 .1 3.1 . Che cking Ne t wo rk Acce ss fo r Inco m ing HT T PS and HT T PS Using t he
Co m m and Line

4 05

Deployment G uide

To check what the firewall is configured to allow, using the command line, issue the following
command as ro o t :
~]# less /et c/sysco n f ig /syst em- co n f ig - f irewall
# Configuration file for system-config-firewall
--enabled
--service=ssh
In this example taken from a default installation, the firewall is enabled but H T T P and H T T PS have
not been allowed to pass through.
Once the default port for H T T P is enabled, the following line appears as output in addition to the
lines shown above:
--port=80:tcp
To check if the firewall is currently allowing incoming H T T P traffic for clients, issue the following
command as ro o t :
~]# ip t ab les - L - n | g rep ' t cp .*80'
ACCEPT tcp -- 0.0.0.0/0
0.0.0.0/0

state NEW tcp dpt:80

Once the default port for H T T PS is enabled, the following line appears as output in addition to the
lines shown above:
--port=443:tcp
To check if the firewall is currently allowing incoming H T T PS traffic for clients, issue the following
command as ro o t :
~]# ip t ab les - L - n | g rep ' t cp .*4 4 3'
ACCEPT tcp -- 0.0.0.0/0
0.0.0.0/0

state NEW tcp dpt:443

17.1.14 . Addit ional Resources
To learn more about the Apache HTTP Server, see the following resources.

Inst alle d Do cum e nt at io n
h t t p d ( 8) — The manual page for the h t t p d service containing the complete list of its commandline options.
g en key( 1) — The manual page for g en key utility, provided by the crypto-utils package.

Inst allable Do cum e nt at io n
http://localhost/manual/ — The official documentation for the Apache HTTP Server with the full
description of its directives and available modules. Note that in order to access this
documentation, you must have the httpd-manual package installed, and the web server must be
running.
Before accessing the documentation, issue the following commands as ro o t :

4 06

⁠Chapt er 1 7 . Web Servers

~]# yu m in st all h t t p d - man u al
~]# service h t t p d g racef u l

Online Do cum e nt at io n
http://httpd.apache.org/ — The official website for the Apache HTTP Server with documentation on
all the directives and default modules.
http://www.openssl.org/ — The OpenSSL home page containing further documentation, frequently
asked questions, links to the mailing lists, and other useful resources.

4 07

Deployment G uide

Chapter 18. Mail Servers
Red Hat Enterprise Linux offers many advanced applications to serve and access email. This chapter
describes modern email protocols in use today, and some of the programs designed to send and
receive email.

18.1. Email Prot ocols
Today, email is delivered using a client/server architecture. An email message is created using a mail
client program. This program then sends the message to a server. The server then forwards the
message to the recipient's email server, where the message is then supplied to the recipient's email
client.
To enable this process, a variety of standard network protocols allow different machines, often
running different operating systems and using different email programs, to send and receive email.
The following protocols discussed are the most commonly used in the transfer of email.

18.1.1. Mail T ransport Prot ocols
Mail delivery from a client application to the server, and from an originating server to the destination
server, is handled by the Simple Mail Transfer Protocol (SMTP).

1 8 .1 .1 .1 . SMT P
The primary purpose of SMTP is to transfer email between mail servers. However, it is critical for email
clients as well. To send email, the client sends the message to an outgoing mail server, which in turn
contacts the destination mail server for delivery. For this reason, it is necessary to specify an SMTP
server when configuring an email client.
Under Red Hat Enterprise Linux, a user can configure an SMTP server on the local machine to
handle mail delivery. However, it is also possible to configure remote SMTP servers for outgoing mail.
One important point to make about the SMTP protocol is that it does not require authentication. This
allows anyone on the Internet to send email to anyone else or even to large groups of people. It is
this characteristic of SMTP that makes junk email or spam possible. Imposing relay restrictions limits
random users on the Internet from sending email through your SMTP server, to other servers on the
internet. Servers that do not impose such restrictions are called open relay servers.
Red Hat Enterprise Linux provides the Postfix and Sendmail SMTP programs.

18.1.2. Mail Access Prot ocols
There are two primary protocols used by email client applications to retrieve email from mail servers:
the Post Office Protocol (POP) and the Internet Message Access Protocol (IMAP).

1 8 .1 .2 .1 . POP
The default POP server under Red Hat Enterprise Linux is D o veco t and is provided by the dovecot
package.

4 08

⁠Chapt er 1 8 . Mail Servers

Installing the dovecot package
In order to use D o veco t , first ensure the dovecot package is installed on your system by
running, as ro o t :
~]# yu m in st all d o veco t
For more information on installing packages with Yum, see Section 7.2.4, “ Installing
Packages” .

When using a PO P server, email messages are downloaded by email client applications. By default,
most PO P email clients are automatically configured to delete the message on the email server after it
has been successfully transferred, however this setting usually can be changed.
PO P is fully compatible with important Internet messaging standards, such as Multipurpose Internet
Mail Extensions (MIME), which allow for email attachments.
PO P works best for users who have one system on which to read email. It also works well for users
who do not have a persistent connection to the Internet or the network containing the mail server.
Unfortunately for those with slow network connections, PO P requires client programs upon
authentication to download the entire content of each message. This can take a long time if any
messages have large attachments.
The most current version of the standard PO P protocol is PO P3.
There are, however, a variety of lesser-used PO P protocol variants:
APOP — PO P3 with MD 5 authentication. An encoded hash of the user's password is sent from
the email client to the server rather than sending an unencrypted password.
KPOP — PO P3 with Kerberos authentication.
RPOP — PO P3 with R PO P authentication. This uses a per-user ID , similar to a password, to
authenticate POP requests. However, this ID is not encrypted, so R PO P is no more secure than
standard PO P.
For added security, it is possible to use Secure Socket Layer (SSL) encryption for client authentication
and data transfer sessions. This can be enabled by using the p o p 3s service, or by using the
st u n n el application. For more information on securing email communication, see Section 18.5.1,
“ Securing Communication” .

1 8 .1 .2 .2 . IMAP
The default IMAP server under Red Hat Enterprise Linux is D o veco t and is provided by the dovecot
package. See Section 18.1.2.1, “ POP” for information on how to install D o veco t .
When using an IMAP mail server, email messages remain on the server where users can read or
delete them. IMAP also allows client applications to create, rename, or delete mail directories on the
server to organize and store email.
IMAP is particularly useful for users who access their email using multiple machines. The protocol is
also convenient for users connecting to the mail server via a slow connection, because only the
email header information is downloaded for messages until opened, saving bandwidth. The user
also has the ability to delete messages without viewing or downloading them.

4 09

Deployment G uide

For convenience, IMAP client applications are capable of caching copies of messages locally, so
the user can browse previously read messages when not directly connected to the IMAP server.
IMAP, like PO P, is fully compatible with important Internet messaging standards, such as MIME,
which allow for email attachments.
For added security, it is possible to use SSL encryption for client authentication and data transfer
sessions. This can be enabled by using the imap s service, or by using the st u n n el program. For
more information on securing email communication, see Section 18.5.1, “ Securing Communication” .
Other free, as well as commercial, IMAP clients and servers are available, many of which extend the
IMAP protocol and provide additional functionality.

1 8 .1 .2 .3. Do ve co t
The imap - lo g in and p o p 3- lo g in processes which implement the IMAP and PO P3 protocols are
spawned by the master d o veco t daemon included in the dovecot package. The use of IMAP and
PO P is configured through the /et c/d o veco t /d o veco t .co n f configuration file; by default d o veco t
runs IMAP and PO P3 together with their secure versions using SSL. To configure d o veco t to use
PO P, complete the following steps:
1. Edit the /et c/d o veco t /d o veco t .co n f configuration file to make sure the p ro t o co ls variable
is uncommented (remove the hash sign (#) at the beginning of the line) and contains the
p o p 3 argument. For example:
protocols = imap pop3 lmtp
When the p ro t o co ls variable is left commented out, d o veco t will use the default values as
described above.
2. Make the change operational for the current session by running the following command:
~]# service d o veco t rest art
3. Make the change operational after the next reboot by running the command:
~]# ch kco n f ig d o veco t o n

T he dovecot service starts the POP3 server
Please note that d o veco t only reports that it started the IMAP server, but also starts
the PO P3 server.
Unlike SMT P, both IMAP and PO P3 require connecting clients to authenticate using a user name
and password. By default, passwords for both protocols are passed over the network unencrypted.
To configure SSL on d o veco t :
Edit the /et c/d o veco t /co n f .d /10- ssl.co n f configuration to make sure the ssl_cip h er_list
variable is uncommented, and append :!SSLv3:
ssl_cipher_list = ALL:!LOW:!SSLv2:!EXP:!aNULL:!SSLv3
These values ensure that d o veco t avoids SSL versions 2 and also 3, which are both known to

4 10

⁠Chapt er 1 8 . Mail Servers

be insecure. This is due to the vulnerability described in POODLE: SSLv3 vulnerability (CVE-20143566). See Resolution for POODLE SSL 3.0 vulnerability (CVE-2014-3566) in Postfix and Dovecot for
details.
Edit the /et c/p ki/d o veco t /d o veco t - o p en ssl.cn f configuration file as you prefer. However, in a
typical installation, this file does not require modification.
Rename, move or delete the files /et c/p ki/d o veco t /cert s/d o veco t .p em and
/et c/p ki/d o veco t /p rivat e/d o veco t .p em.
Execute the /u sr/lib exec/d o veco t /mkcert .sh script which creates the d o veco t self signed
certificates. These certificates are copied in the /et c/p ki/d o veco t /cert s and
/et c/p ki/d o veco t /p rivat e directories. To implement the changes, restart d o veco t :
~]# service d o veco t rest art
More details on d o veco t can be found online at http://www.dovecot.org.

18.2. Email Program Classificat ions
In general, all email applications fall into at least one of three classifications. Each classification
plays a specific role in the process of moving and managing email messages. While most users are
only aware of the specific email program they use to receive and send messages, each one is
important for ensuring that email arrives at the correct destination.

18.2.1. Mail T ransport Agent
A Mail Transport Agent (MTA) transports email messages between hosts using SMT P. A message may
involve several MTAs as it moves to its intended destination.
While the delivery of messages between machines may seem rather straightforward, the entire
process of deciding if a particular MTA can or should accept a message for delivery is quite
complicated. In addition, due to problems from spam, use of a particular MTA is usually restricted by
the MTA's configuration or the access configuration for the network on which the MTA resides.
Many modern email client programs can act as an MTA when sending email. However, this action
should not be confused with the role of a true MTA. The sole reason email client programs are
capable of sending email like an MTA is because the host running the application does not have its
own MTA. This is particularly true for email client programs on non-UNIX-based operating systems.
However, these client programs only send outbound messages to an MTA they are authorized to use
and do not directly deliver the message to the intended recipient's email server.
Since Red Hat Enterprise Linux offers two MTAs, Postfix and Sendmail, email client programs are often
not required to act as an MTA. Red Hat Enterprise Linux also includes a special purpose MTA called
Fetchmail.
For more information on Postfix, Sendmail, and Fetchmail, see Section 18.3, “ Mail Transport Agents” .

18.2.2. Mail Delivery Agent
A Mail Delivery Agent (MDA) is invoked by the MTA to file incoming email in the proper user's mailbox.
In many cases, the MD A is actually a Local Delivery Agent (LDA), such as mail or Procmail.
Any program that actually handles a message for delivery to the point where it can be read by an
email client application can be considered an MD A. For this reason, some MTAs (such as Sendmail
and Postfix) can fill the role of an MD A when they append new email messages to a local user's mail

4 11

Deployment G uide

spool file. In general, MD As do not transport messages between systems nor do they provide a user
interface; MD As distribute and sort messages on the local machine for an email client application to
access.

18.2.3. Mail User Agent
A Mail User Agent (MUA) is synonymous with an email client application. An MUA is a program that, at
a minimum, allows a user to read and compose email messages. Many MUAs are capable of
retrieving messages via the PO P or IMAP protocols, setting up mailboxes to store messages, and
sending outbound messages to an MTA.
MUAs may be graphical, such as Evo lu t io n , or have simple text-based interfaces, such as p in e.

18.3. Mail T ransport Agent s
Red Hat Enterprise Linux offers two primary MTAs: Postfix and Sendmail. Postfix is configured as the
default MTA, although it is easy to switch the default MTA to Sendmail. To switch the default MTA to
Sendmail, you can either uninstall Postfix or use the following command to switch to Sendmail:
~ ] # alt ern at ives - - co n f ig mt a
You can also use a command in the following format to enable or disable the desired service:
chkconfig service_name on | off

18.3.1. Post fix
Originally developed at IBM by security expert and programmer Wietse Venema, Postfix is a
Sendmail-compatible MTA that is designed to be secure, fast, and easy to configure.
To improve security, Postfix uses a modular design, where small processes with limited privileges are
launched by a master daemon. The smaller, less privileged processes perform very specific tasks
related to the various stages of mail delivery and run in a changed root environment to limit the
effects of attacks.
Configuring Postfix to accept network connections from hosts other than the local computer takes
only a few minor changes in its configuration file. Yet for those with more complex needs, Postfix
provides a variety of configuration options, as well as third party add-ons that make it a very
versatile and full-featured MTA.
The configuration files for Postfix are human readable and support upward of 250 directives. Unlike
Sendmail, no macro processing is required for changes to take effect and the majority of the most
commonly used options are described in the heavily commented files.

1 8 .3.1 .1 . T he De fault Po st fix Inst allat io n
The Postfix executable is /u sr/sb in /p o st f ix. This daemon launches all related processes needed to
handle mail delivery.
Postfix stores its configuration files in the /et c/p o st f ix/ directory. The following is a list of the more
commonly used files:
access — Used for access control, this file specifies which hosts are allowed to connect to
Postfix.

4 12

⁠Chapt er 1 8 . Mail Servers

main .cf — The global Postfix configuration file. The majority of configuration options are
specified in this file.
mast er.cf — Specifies how Postfix interacts with various processes to accomplish mail delivery.
t ran sp o rt — Maps email addresses to relay hosts.
The aliases file can be found in the /et c/ directory. This file is shared between Postfix and Sendmail.
It is a configurable list required by the mail protocol that describes user ID aliases.

Configuring Postfix as a server for other clients
The default /et c/p o st f ix/main .cf file does not allow Postfix to accept network connections
from a host other than the local computer. For instructions on configuring Postfix as a server
for other clients, see Section 18.3.1.2, “ Basic Postfix Configuration” .
Restart the p o st f ix service after changing any options in the configuration files under the
/et c/p o st f ix directory in order for those changes to take effect:
~]# service p o st f ix rest art

1 8 .3.1 .2 . Basic Po st fix Co nfigurat io n
By default, Postfix does not accept network connections from any host other than the local host.
Perform the following steps as ro o t to enable mail delivery for other hosts on the network:
Edit the /et c/p o st f ix/main .cf file with a text editor, such as vi.
Uncomment the myd o main line by removing the hash sign (#), and replace domain.tld with the
domain the mail server is servicing, such as examp le.co m.
Uncomment the myo rig in = $myd o main line.
Uncomment the myh o st n ame line, and replace host.domain.tld with the host name for the
machine.
Uncomment the myd est in at io n = $myh o st n ame, lo calh o st .$myd o main line.
Uncomment the myn et wo rks line, and replace 168.100.189.0/28 with a valid network setting for
hosts that can connect to the server.
Uncomment the in et _in t erf aces = all line.
Comment the in et _in t erf aces = lo calh o st line.
Restart the p o st f ix service.
Once these steps are complete, the host accepts outside emails for delivery.
Postfix has a large assortment of configuration options. One of the best ways to learn how to
configure Postfix is to read the comments within the /et c/p o st f ix/main .cf configuration file.
Additional resources including information about Postfix configuration, SpamAssassin integration,
or detailed descriptions of the /et c/p o st f ix/main .cf parameters are available online at
http://www.postfix.org/.
18.3.1.2.1. C o n f ig u rin g Po st f ix t o U se T ran sp o rt Layer Secu rit y

4 13

Deployment G uide

Configuring postfix to use transport layer security (TLS) is described in the Red Hat Knowledgebase
solution How to configure postfix with TLS?

Important
D ue to the vulnerability described in Resolution for POODLE SSL 3.0 vulnerability (CVE-20143566) in Postfix and Dovecot, Red Hat recommends disabling SSL and using only T LSv1.1 or
T LSv1.2. Backwards compatibility can be achieved using T LSv1.0. Many products Red Hat
supports have the ability to use SSLv2 or SSLv3 protocols, or enable them by default.
However, the use of SSLv2 or SSLv3 is now strongly recommended against.

1 8 .3.1 .3. Using Po st fix wit h LDAP
Postfix can use an LD AP directory as a source for various lookup tables (e.g.: aliases, virt u al,
can o n ical, etc.). This allows LD AP to store hierarchical user information and Postfix to only be
given the result of LD AP queries when needed. By not storing this information locally, administrators
can easily maintain it.
18.3.1.3.1. T h e /et c/aliases lo o ku p examp le
The following is a basic example for using LD AP to look up the /et c/aliases file. Make sure your
/et c/p o st f ix/main .cf file contains the following:
alias_maps = hash:/etc/aliases, ldap:/etc/postfix/ldap-aliases.cf
Create a /et c/p o st f ix/ld ap - aliases.cf file if you do not have one already and make sure it contains
the following:
server_host = ldap.example.com
search_base = dc=example, dc=com
where ldap.example.com, example, and com are parameters that need to be replaced with
specification of an existing available LD AP server.

T he /etc/postfix/ldap-aliases.cf file
The /et c/p o st f ix/ld ap - aliases.cf file can specify various parameters, including parameters
that enable LD AP SSL and ST AR T T LS. For more information, see the ld ap _t ab le( 5) man
page.
For more information on LD AP, see Section 19.1, “ OpenLD AP” .

18.3.2. Sendmail
Sendmail's core purpose, like other MTAs, is to safely transfer email among hosts, usually using the
SMT P protocol. However, Sendmail is highly configurable, allowing control over almost every aspect
of how email is handled, including the protocol used. Many system administrators elect to use
Sendmail as their MTA due to its power and scalability.

1 8 .3.2 .1 . Purpo se and Lim it at io ns

4 14

⁠Chapt er 1 8 . Mail Servers

It is important to be aware of what Sendmail is and what it can do, as opposed to what it is not. In
these days of monolithic applications that fulfill multiple roles, Sendmail may seem like the only
application needed to run an email server within an organization. Technically, this is true, as
Sendmail can spool mail to each users' directory and deliver outbound mail for users. However, most
users actually require much more than simple email delivery. Users usually want to interact with their
email using an MUA, that uses PO P or IMAP, to download their messages to their local machine. Or,
they may prefer a Web interface to gain access to their mailbox. These other applications can work in
conjunction with Sendmail, but they actually exist for different reasons and can operate separately
from one another.
It is beyond the scope of this section to go into all that Sendmail should or could be configured to
do. With literally hundreds of different options and rule sets, entire volumes have been dedicated to
helping explain everything that can be done and how to fix things that go wrong. See the
Section 18.6, “ Additional Resources” for a list of Sendmail resources.
This section reviews the files installed with Sendmail by default and reviews basic configuration
changes, including how to stop unwanted email (spam) and how to extend Sendmail with the
Lightweight Directory Access Protocol (LDAP).

1 8 .3.2 .2 . T he De fault Se ndm ail Inst allat io n
In order to use Sendmail, first ensure the sendmail package is installed on your system by running, as
ro o t :
~]# yu m in st all sen d mail
In order to configure Sendmail, ensure the sendmail-cf package is installed on your system by
running, as ro o t :
~]# yu m in st all sen d mail- cf
For more information on installing packages with Yum, see Section 7.2.4, “ Installing Packages” .
Before using Sendmail, the default MTA has to be switched from Postfix. For more information how to
switch the default MTA see Section 18.3, “ Mail Transport Agents” .
The Sendmail executable is /u sr/sb in /sen d mail.
Sendmail's lengthy and detailed configuration file is /et c/mail/sen d mail.cf . Avoid editing the
sen d mail.cf file directly. To make configuration changes to Sendmail, edit the
/et c/mail/sen d mail.mc file, back up the original /et c/mail/sen d mail.cf file, and use the following
alternatives to generate a new configuration file:
Use the included makefile in /et c/mail/ to create a new /et c/mail/sen d mail.cf configuration file:
~]# make all - C /et c/mail/
All other generated files in /et c/mail (db files) will be regenerated if needed. The old makemap
commands are still usable. The make command is automatically used whenever you start or
restart the sen d mail service.
Alternatively you may use the m4 macro processor to create a new /et c/mail/sen d mail.cf . The
m4 macro processor is not installed by default. Before using it to create /et c/mail/sen d mail.cf ,
install the m4 package as root:
~]# yu m in st all m4

4 15

Deployment G uide

More information on configuring Sendmail can be found in Section 18.3.2.3, “ Common Sendmail
Configuration Changes” .
Various Sendmail configuration files are installed in the /et c/mail/ directory including:
access — Specifies which systems can use Sendmail for outbound email.
d o main t ab le — Specifies domain name mapping.
lo cal- h o st - n ames — Specifies aliases for the host.
mailert ab le — Specifies instructions that override routing for particular domains.
virt u sert ab le — Specifies a domain-specific form of aliasing, allowing multiple virtual domains
to be hosted on one machine.
Several of the configuration files in /et c/mail/, such as access, d o main t ab le, mailert ab le and
virt u sert ab le, must actually store their information in database files before Sendmail can use any
configuration changes. To include any changes made to these configurations in their database files,
run the following command, as ro o t :
~]# makemap h ash /et c/mail/ < /et c/mail/
where  represents the name of the configuration file to be updated. You may also restart the
sen d mail service for the changes to take effect by running:
~]# service sen d mail rest art
For example, to have all emails addressed to the examp le.co m domain delivered to b o b @ o t h erexamp le.co m, add the following line to the virt u sert ab le file:
@ examp le.co m b o b @ o t h er- examp le.co m
To finalize the change, the virt u sert ab le.d b file must be updated:
~]# makemap h ash /et c/mail/virt u sert ab le < /et c/mail/virt u sert ab le
Sendmail will create an updated virt u sert ab le.d b file containing the new configuration.

1 8 .3.2 .3. Co m m o n Se ndm ail Co nfigurat io n Change s
When altering the Sendmail configuration file, it is best not to edit an existing file, but to generate an
entirely new /et c/mail/sen d mail.cf file.

Backup the sendmail.cf file before changing its content
Before replacing or making any changes to the sen d mail.cf file, create a backup copy.
To add the desired functionality to Sendmail, edit the /et c/mail/sen d mail.mc file as root. Once you
are finished, restart the sen d mail service and, if the m4 package is installed, the m4 macro
processor will automatically generate a new sen d mail.cf configuration file:
~]# service sen d mail rest art

4 16

⁠Chapt er 1 8 . Mail Servers

Configuring Sendmail as a server for other clients
The default sen d mail.cf file does not allow Sendmail to accept network connections from any
host other than the local computer. To configure Sendmail as a server for other clients, edit the
/et c/mail/sen d mail.mc file, and either change the address specified in the Ad d r= option of
the D AEMO N _O PT IO N S directive from 127.0.0.1 to the IP address of an active network
device or comment out the D AEMO N _O PT IO N S directive all together by placing d n l at the
beginning of the line. When finished, regenerate /et c/mail/sen d mail.cf by restarting the
service
~]# service sen d mail rest art

The default configuration in Red Hat Enterprise Linux works for most SMT P-only sites. However, it
does not work for UUCP (UNIX-to-UNIX Copy Protocol) sites. If using UUCP mail transfers, the
/et c/mail/sen d mail.mc file must be reconfigured and a new /et c/mail/sen d mail.cf file must be
generated.
Consult the /u sr/sh are/sen d mail- cf /R EAD ME file before editing any files in the directories under
the /u sr/sh are/sen d mail- cf directory, as they can affect the future configuration of the
/et c/mail/sen d mail.cf file.

1 8 .3.2 .4 . Masque rading
One common Sendmail configuration is to have a single machine act as a mail gateway for all
machines on the network. For example, a company may want to have a machine called
mail.examp le.co m that handles all of their email and assigns a consistent return address to all
outgoing mail.
In this situation, the Sendmail server must masquerade the machine names on the company network
so that their return address is u ser@ examp le.co m instead of u ser@ h o st .examp le.co m.
To do this, add the following lines to /et c/mail/sen d mail.mc:
FEATURE(always_add_domain)dnl
FEATURE(`masquerade_entire_domain')dnl
FEATURE(`masquerade_envelope')dnl
FEATURE(`allmasquerade')dnl
MASQUERAD E_AS(`example.com.')dnl
MASQUERAD E_D OMAIN(`example.com.')dnl
MASQUERAD E_AS(example.com)dnl
After generating a new sen d mail.cf file using the m4 macro processor, this configuration makes all
mail from inside the network appear as if it were sent from examp le.co m.

1 8 .3.2 .5 . St o pping Spam
Email spam can be defined as unnecessary and unwanted email received by a user who never
requested the communication. It is a disruptive, costly, and widespread abuse of Internet
communication standards.

4 17

Deployment G uide

Sendmail makes it relatively easy to block new spamming techniques being employed to send junk
email. It even blocks many of the more usual spamming methods by default. Main anti-spam features
available in sendmail are header checks, relaying denial (default from version 8.9), access database and
sender information checks.
For example, forwarding of SMT P messages, also called relaying, has been disabled by default
since Sendmail version 8.9. Before this change occurred, Sendmail directed the mail host (x.ed u ) to
accept messages from one party (y.co m) and sent them to a different party (z .n et ). Now, however,
Sendmail must be configured to permit any domain to relay mail through the server. To configure
relay domains, edit the /et c/mail/relay- d o main s file and restart Sendmail
~]# service sen d mail rest art
However users can also be sent spam from from servers on the Internet. In these instances,
Sendmail's access control features available through the /et c/mail/access file can be used to
prevent connections from unwanted hosts. The following example illustrates how this file can be used
to both block and specifically allow access to the Sendmail server:
b ad sp ammer.co m ER R O R :550 "G o away an d d o n o t sp am u s an ymo re"
t u x.b ad sp ammer.co m O K 10.0 R ELAY
This example shows that any email sent from b ad sp ammer.co m is blocked with a 550 RFC-821
compliant error code, with a message sent back. Email sent from the t u x.b ad sp ammer.co m subdomain, is accepted. The last line shows that any email sent from the 10.0.*.* network can be relayed
through the mail server.
Because the /et c/mail/access.d b file is a database, use the makemap command to update any
changes. D o this using the following command as ro o t :
~]# makemap h ash /et c/mail/access < /et c/mail/access
Message header analysis allows you to reject mail based on header contents. SMT P servers store
information about an email's journey in the message header. As the message travels from one MTA to
another, each puts in a R eceived header above all the other R eceived headers. It is important to
note that this information may be altered by spammers.
The above examples only represent a small part of what Sendmail can do in terms of allowing or
blocking access. See the /u sr/sh are/sen d mail- cf /R EAD ME file for more information and examples.
Since Sendmail calls the Procmail MD A when delivering mail, it is also possible to use a spam
filtering program, such as SpamAssassin, to identify and file spam for users. See Section 18.4.2.6,
“ Spam Filters” for more information about using SpamAssassin.

1 8 .3.2 .6 . Using Se ndm ail wit h LDAP
Using LD AP is a very quick and powerful way to find specific information about a particular user
from a much larger group. For example, an LD AP server can be used to look up a particular email
address from a common corporate directory by the user's last name. In this kind of implementation,
LD AP is largely separate from Sendmail, with LD AP storing the hierarchical user information and
Sendmail only being given the result of LD AP queries in pre-addressed email messages.
However, Sendmail supports a much greater integration with LD AP, where it uses LD AP to replace
separately maintained files, such as /et c/aliases and /et c/mail/virt u sert ab les, on different mail
servers that work together to support a medium- to enterprise-level organization. In short, LD AP
abstracts the mail routing level from Sendmail and its separate configuration files to a powerful
LD AP cluster that can be leveraged by many different applications.

4 18

⁠Chapt er 1 8 . Mail Servers

The current version of Sendmail contains support for LD AP. To extend the Sendmail server using
LD AP, first get an LD AP server, such as O p en LD AP, running and properly configured. Then edit
the /et c/mail/sen d mail.mc to include the following:
LD APROUTE_D OMAIN('yourdomain.com')dnl
FEATURE('ldap_routing')dnl

Advanced configuration
This is only for a very basic configuration of Sendmail with LD AP. The configuration can differ
greatly from this depending on the implementation of LD AP, especially when configuring
several Sendmail machines to use a common LD AP server.
Consult /u sr/sh are/sen d mail- cf /R EAD ME for detailed LD AP routing configuration
instructions and examples.

Next, recreate the /et c/mail/sen d mail.cf file by running the m4 macro processor and again
restarting Sendmail. See Section 18.3.2.3, “ Common Sendmail Configuration Changes” for
instructions.
For more information on LD AP, see Section 19.1, “ OpenLD AP” .

18.3.3. Fet chmail
Fetchmail is an MTA which retrieves email from remote servers and delivers it to the local MTA. Many
users appreciate the ability to separate the process of downloading their messages located on a
remote server from the process of reading and organizing their email in an MUA. D esigned with the
needs of dial-up users in mind, Fetchmail connects and quickly downloads all of the email messages
to the mail spool file using any number of protocols, including PO P3 and IMAP. It can even forward
email messages to an SMT P server, if necessary.

Installing the fetchmail package
In order to use Fet ch mail, first ensure the fetchmail package is installed on your system by
running, as ro o t :
~]# yu m in st all f et ch mail
For more information on installing packages with Yum, see Section 7.2.4, “ Installing
Packages” .

Fetchmail is configured for each user through the use of a .f et ch mailrc file in the user's home
directory. If it does not already exist, create the .f et ch mailrc file in your home directory
Using preferences in the .f et ch mailrc file, Fetchmail checks for email on a remote server and
downloads it. It then delivers it to port 25 on the local machine, using the local MTA to place the email
in the correct user's spool file. If Procmail is available, it is launched to filter the email and place it in
a mailbox so that it can be read by an MUA.

4 19

Deployment G uide

1 8 .3.3.1 . Fe t chm ail Co nfigurat io n Opt io ns
Although it is possible to pass all necessary options on the command line to check for email on a
remote server when executing Fetchmail, using a .f et ch mailrc file is much easier. Place any desired
configuration options in the .f et ch mailrc file for those options to be used each time the f et ch mail
command is issued. It is possible to override these at the time Fetchmail is run by specifying that
option on the command line.
A user's .f et ch mailrc file contains three classes of configuration options:
global options — Gives Fetchmail instructions that control the operation of the program or provide
settings for every connection that checks for email.
server options — Specifies necessary information about the server being polled, such as the host
name, as well as preferences for specific email servers, such as the port to check or number of
seconds to wait before timing out. These options affect every user using that server.
user options — Contains information, such as user name and password, necessary to authenticate
and check for email using a specified email server.
Global options appear at the top of the .f et ch mailrc file, followed by one or more server options,
each of which designate a different email server that Fetchmail should check. User options follow
server options for each user account checking that email server. Like server options, multiple user
options may be specified for use with a particular server as well as to check multiple email accounts
on the same server.
Server options are called into service in the .f et ch mailrc file by the use of a special option verb,
p o ll or skip , that precedes any of the server information. The p o ll action tells Fetchmail to use this
server option when it is run, which checks for email using the specified user options. Any server
options after a skip action, however, are not checked unless this server's host name is specified
when Fetchmail is invoked. The skip option is useful when testing configurations in the
.f et ch mailrc file because it only checks skipped servers when specifically invoked, and does not
affect any currently working configurations.
The following is an example of a .f et ch mailrc file:
set postmaster " user1"
set bouncemail
poll pop.domain.com proto pop3
user 'user1' there with password 'secret' is user1 here
poll mail.domain2.com
user 'user5' there with password 'secret2' is user1 here
user 'user7' there with password 'secret3' is user1 here
In this example, the global options specify that the user is sent email as a last resort (p o st mast er
option) and all email errors are sent to the postmaster instead of the sender (b o u n cemail option).
The set action tells Fetchmail that this line contains a global option. Then, two email servers are
specified, one set to check using PO P3, the other for trying various protocols to find one that works.
Two users are checked using the second server option, but all email found for any user is sent to
u ser1's mail spool. This allows multiple mailboxes to be checked on multiple servers, while
appearing in a single MUA inbox. Each user's specific information begins with the u ser action.

4 20

⁠Chapt er 1 8 . Mail Servers

Omitting the password from the configuration
Users are not required to place their password in the .f et ch mailrc file. Omitting the wit h
p asswo rd ' 
' section causes Fetchmail to ask for a password when it is
launched.
Fetchmail has numerous global, server, and local options. Many of these options are rarely used or
only apply to very specific situations. The f et ch mail man page explains each option in detail, but
the most common ones are listed in the following three sections.

1 8 .3.3.2 . Glo bal Opt io ns
Each global option should be placed on a single line after a set action.
d aemo n seconds — Specifies daemon-mode, where Fetchmail stays in the background. Replace
seconds with the number of seconds Fetchmail is to wait before polling the server.
p o st mast er — Specifies a local user to send mail to in case of delivery problems.
syslo g — Specifies the log file for errors and status messages. By default, this is
/var/lo g /maillo g .

1 8 .3.3.3. Se rve r Opt io ns
Server options must be placed on their own line in .f et ch mailrc after a p o ll or skip action.
au t h auth-type — Replace auth-type with the type of authentication to be used. By default,
p asswo rd authentication is used, but some protocols support other types of authentication,
including kerb ero s_v5, kerb ero s_v4 , and ssh . If the an y authentication type is used,
Fetchmail first tries methods that do not require a password, then methods that mask the
password, and finally attempts to send the password unencrypted to authenticate to the server.
in t erval number — Polls the specified server every number of times that it checks for email on
all configured servers. This option is generally used for email servers where the user rarely
receives messages.
p o rt port-number — Replace port-number with the port number. This value overrides the default
port number for the specified protocol.
p ro t o protocol — Replace protocol with the protocol, such as p o p 3 or imap , to use when
checking for messages on the server.
t imeo u t seconds — Replace seconds with the number of seconds of server inactivity after which
Fetchmail gives up on a connection attempt. If this value is not set, a default of 300 seconds is
used.

1 8 .3.3.4 . Use r Opt io ns
User options may be placed on their own lines beneath a server option or on the same line as the
server option. In either case, the defined options must follow the u ser option (defined below).
f et ch all — Orders Fetchmail to download all messages in the queue, including messages that
have already been viewed. By default, Fetchmail only pulls down new messages.
f et ch limit number — Replace number with the number of messages to be retrieved before
stopping.

4 21

Deployment G uide

f lu sh — D eletes all previously viewed messages in the queue before retrieving new messages.
limit max-number-bytes — Replace max-number-bytes with the maximum size in bytes that
messages are allowed to be when retrieved by Fetchmail. This option is useful with slow network
links, when a large message takes too long to download.
p asswo rd ' password' — Replace password with the user's password.
p reco n n ect "command" — Replace command with a command to be executed before retrieving
messages for the user.
p o st co n n ect "command" — Replace command with a command to be executed after retrieving
messages for the user.
ssl — Activates SSL encryption. At the time of writing, the default action is to use the best
available from SSL2, SSL3, SSL23, T LS1, T LS1.1 and T LS1.2. Note that SSL2 is considered
obsolete and due to the POOD LE: SSLv3 vulnerability (CVE-2014-3566), SSLv3 should not be
used. However there is no way to force the use of TLS1 or newer, therefore ensure the mail server
being connected to is configured n o t to use SSLv2 and SSLv3. Use st u n n el where the server
cannot be configured n o t to use SSLv2 and SSLv3.
sslp ro t o — D efines allowed SSL or TLS protocols. Possible values are SSL2, SSL3, SSL23,
and T LS1. The default value, if sslp ro t o is omitted, unset, or set to an invalid value, is SSL23.
The default action is to use the best from SSLv2, SSLv3, T LSv1, T LS1.1 and T LS1.2. Note that
setting any other value for SSL or TLS will disable all the other protocols. D ue to the POOD LE:
SSLv3 vulnerability (CVE-2014-3566), it is recommend to omit this option, or set it to SSLv23, and
configure the corresponding mail server n o t to use SSLv2 and SSLv3. Use st u n n el where the
server cannot be configured n o t to use SSLv2 and SSLv3.
u ser "username" — Replace username with the username used by Fetchmail to retrieve
messages. This option must precede all other user options.

1 8 .3.3.5 . Fe t chm ail Co m m and Opt io ns
Most Fetchmail options used on the command line when executing the f et ch mail command mirror
the .f et ch mailrc configuration options. In this way, Fetchmail may be used with or without a
configuration file. These options are not used on the command line by most users because it is
easier to leave them in the .f et ch mailrc file.
There may be times when it is desirable to run the f et ch mail command with other options for a
particular purpose. It is possible to issue command options to temporarily override a .f et ch mailrc
setting that is causing an error, as any options specified at the command line override configuration
file options.

1 8 .3.3.6 . Info rm at io nal o r De bugging Opt io ns
Certain options used after the f et ch mail command can supply important information.
- - co n f ig d u mp — D isplays every possible option based on information from .f et ch mailrc and
Fetchmail defaults. No email is retrieved for any users when using this option.
- s — Executes Fetchmail in silent mode, preventing any messages, other than errors, from
appearing after the f et ch mail command.
- v — Executes Fetchmail in verbose mode, displaying every communication between Fetchmail
and remote email servers.

4 22

⁠Chapt er 1 8 . Mail Servers

- V — D isplays detailed version information, lists its global options, and shows settings to be used
with each user, including the email protocol and authentication method. No email is retrieved for
any users when using this option.

1 8 .3.3.7 . Spe cial Opt io ns
These options are occasionally useful for overriding defaults often found in the .f et ch mailrc file.
- a — Fetchmail downloads all messages from the remote email server, whether new or previously
viewed. By default, Fetchmail only downloads new messages.
- k — Fetchmail leaves the messages on the remote email server after downloading them. This
option overrides the default behavior of deleting messages after downloading them.
- l max-number-bytes — Fetchmail does not download any messages over a particular size and
leaves them on the remote email server.
- - q u it — Quits the Fetchmail daemon process.
More commands and .f et ch mailrc options can be found in the f et ch mail man page.

18.3.4 . Mail T ransport Agent (MT A) Configurat ion
A Mail Transport Agent (MTA) is essential for sending email. A Mail User Agent (MUA) such as
Evo lu t io n , T h u n d erb ird , or Mu t t , is used to read and compose email. When a user sends an
email from an MUA, the message is handed off to the MTA, which sends the message through a series
of MTAs until it reaches its destination.
Even if a user does not plan to send email from the system, some automated tasks or system
programs might use the /b in /mail command to send email containing log messages to the ro o t user
of the local system.
Red Hat Enterprise Linux 6 provides two MTAs: Postfix and Sendmail. If both are installed, Postfix is
the default MTA.

18.4 . Mail Delivery Agent s
Red Hat Enterprise Linux includes two primary MD As, Procmail and mail. Both of the applications
are considered LD As and both move email from the MTA's spool file into the user's mailbox. However,
Procmail provides a robust filtering system.
This section details only Procmail. For information on the mail command, consult its man page
(man mail).
Procmail delivers and filters email as it is placed in the mail spool file of the localhost. It is powerful,
gentle on system resources, and widely used. Procmail can play a critical role in delivering email to
be read by email client applications.
Procmail can be invoked in several different ways. Whenever an MTA places an email into the mail
spool file, Procmail is launched. Procmail then filters and files the email for the MUA and quits.
Alternatively, the MUA can be configured to execute Procmail any time a message is received so that
messages are moved into their correct mailboxes. By default, the presence of /et c/p ro cmailrc or of a
~ /.p ro cmailrc file (also called an rc file) in the user's home directory invokes Procmail whenever an
MTA receives a new message.

4 23

Deployment G uide

By default, no system-wide rc files exist in the /et c/ directory and no .p ro cmailrc files exist in any
user's home directory. Therefore, to use Procmail, each user must construct a .p ro cmailrc file with
specific environment variables and rules.
Whether Procmail acts upon an email message depends upon whether the message matches a
specified set of conditions or recipes in the rc file. If a message matches a recipe, then the email is
placed in a specified file, is deleted, or is otherwise processed.
When Procmail starts, it reads the email message and separates the body from the header
information. Next, Procmail looks for a /et c/p ro cmailrc file and rc files in the /et c/p ro cmailrcs
directory for default, system-wide, Procmail environmental variables and recipes. Procmail then
searches for a .p ro cmailrc file in the user's home directory. Many users also create additional rc
files for Procmail that are referred to within the .p ro cmailrc file in their home directory.

18.4 .1. Procmail Configurat ion
The Procmail configuration file contains important environmental variables. These variables specify
things such as which messages to sort and what to do with the messages that do not match any
recipes.
These environmental variables usually appear at the beginning of the ~ /.p ro cmailrc file in the
following format:
env-variable=" value"
In this example, env-variable is the name of the variable and value defines the variable.
There are many environment variables not used by most Procmail users and many of the more
important environment variables are already defined by a default value. Most of the time, the
following variables are used:
D EFAU LT — Sets the default mailbox where messages that do not match any recipes are placed.
The default D EFAU LT value is the same as $O R G MAIL.
IN C LU D ER C — Specifies additional rc files containing more recipes for messages to be checked
against. This breaks up the Procmail recipe lists into individual files that fulfill different roles, such
as blocking spam and managing email lists, that can then be turned off or on by using comment
characters in the user's ~ /.p ro cmailrc file.
For example, lines in a user's ~ /.p ro cmailrc file may look like this:
MAILD IR=$HOME/Msgs
INCLUD ERC=$MAILD IR/lists.rc
INCLUD ERC=$MAILD IR/spam.rc
To turn off Procmail filtering of email lists but leaving spam control in place, comment out the first
IN C LU D ER C line with a hash sign (#). Note that it uses paths relative to the current directory.
LO C K SLEEP — Sets the amount of time, in seconds, between attempts by Procmail to use a
particular lockfile. The default is 8 seconds.
LO C K T IMEO U T — Sets the amount of time, in seconds, that must pass after a lockfile was last
modified before Procmail assumes that the lockfile is old and can be deleted. The default is 1024
seconds.
LO G FILE — The file to which any Procmail information or error messages are written.

4 24

⁠Chapt er 1 8 . Mail Servers

MAILD IR — Sets the current working directory for Procmail. If set, all other Procmail paths are
relative to this directory.
O R G MAIL — Specifies the original mailbox, or another place to put the messages if they cannot
be placed in the default or recipe-required location.
By default, a value of /var/sp o o l/mail/$LO G N AME is used.
SU SPEN D — Sets the amount of time, in seconds, that Procmail pauses if a necessary resource,
such as swap space, is not available.
SWIT C H R C — Allows a user to specify an external file containing additional Procmail recipes,
much like the IN C LU D ER C option, except that recipe checking is actually stopped on the
referring configuration file and only the recipes on the SWIT C H R C -specified file are used.
VER B O SE — Causes Procmail to log more information. This option is useful for debugging.
Other important environmental variables are pulled from the shell, such as LO G N AME, the login
name; H O ME, the location of the home directory; and SH ELL, the default shell.
A comprehensive explanation of all environments variables, and their default values, is available in
the p ro cmailrc man page.

18.4 .2. Procmail Recipes
New users often find the construction of recipes the most difficult part of learning to use Procmail.
This difficulty is often attributed to recipes matching messages by using regular expressions which are
used to specify qualifications for string matching. However, regular expressions are not very difficult
to construct and even less difficult to understand when read. Additionally, the consistency of the way
Procmail recipes are written, regardless of regular expressions, makes it easy to learn by example. To
see example Procmail recipes, see Section 18.4.2.5, “ Recipe Examples” .
Procmail recipes take the following form:
:0 [flags] [: lockfile-name ]
* [ condition_1_special-condition-character condition_1_regular_expression ]
* [ condition_2_special-condition-character condition-2_regular_expression ]
* [ condition_N_special-condition-character condition-N_regular_expression ]
special-action-character
action-to-perform
The first two characters in a Procmail recipe are a colon and a zero. Various flags can be placed
after the zero to control how Procmail processes the recipe. A colon after the flags section specifies
that a lockfile is created for this message. If a lockfile is created, the name can be specified by
replacing lockfile-name.
A recipe can contain several conditions to match against the message. If it has no conditions, every
message matches the recipe. Regular expressions are placed in some conditions to facilitate
message matching. If multiple conditions are used, they must all match for the action to be performed.
Conditions are checked based on the flags set in the recipe's first line. Optional special characters
placed after the asterisk character (*) can further control the condition.
The action-to-perform argument specifies the action taken when the message matches one of the
conditions. There can only be one action per recipe. In many cases, the name of a mailbox is used
here to direct matching messages into that file, effectively sorting the email. Special action characters
may also be used before the action is specified. See Section 18.4.2.4, “ Special Conditions and
Actions” for more information.

4 25

Deployment G uide

1 8 .4 .2 .1 . De live ring vs. No n-De live ring Re cipe s
The action used if the recipe matches a particular message determines whether it is considered a
delivering or non-delivering recipe. A delivering recipe contains an action that writes the message to a
file, sends the message to another program, or forwards the message to another email address. A
non-delivering recipe covers any other actions, such as a nesting block. A nesting block is a set of
actions, contained in braces { }, that are performed on messages which match the recipe's
conditions. Nesting blocks can be nested inside one another, providing greater control for identifying
and performing actions on messages.
When messages match a delivering recipe, Procmail performs the specified action and stops
comparing the message against any other recipes. Messages that match non-delivering recipes
continue to be compared against other recipes.

1 8 .4 .2 .2 . Flags
Flags are essential to determine how or if a recipe's conditions are compared to a message. The
eg rep utility is used internally for matching of the conditions. The following flags are commonly
used:
A — Specifies that this recipe is only used if the previous recipe without an A or a flag also
matched this message.
a — Specifies that this recipe is only used if the previous recipe with an A or a flag also matched
this message and was successfully completed.
B — Parses the body of the message and looks for matching conditions.
b — Uses the body in any resulting action, such as writing the message to a file or forwarding it.
This is the default behavior.
c — Generates a carbon copy of the email. This is useful with delivering recipes, since the
required action can be performed on the message and a copy of the message can continue being
processed in the rc files.
D — Makes the eg rep comparison case-sensitive. By default, the comparison process is not
case-sensitive.
E — While similar to the A flag, the conditions in the recipe are only compared to the message if
the immediately preceding recipe without an E flag did not match. This is comparable to an else
action.
e — The recipe is compared to the message only if the action specified in the immediately
preceding recipe fails.
f — Uses the pipe as a filter.
H — Parses the header of the message and looks for matching conditions. This is the default
behavior.
h — Uses the header in a resulting action. This is the default behavior.
w — Tells Procmail to wait for the specified filter or program to finish, and reports whether or not it
was successful before considering the message filtered.
W — Is identical to w except that " Program failure" messages are suppressed.
For a detailed list of additional flags, see the p ro cmailrc man page.

4 26

⁠Chapt er 1 8 . Mail Servers

1 8 .4 .2 .3. Spe cifying a Lo cal Lo ckfile
Lockfiles are very useful with Procmail to ensure that more than one process does not try to alter a
message simultaneously. Specify a local lockfile by placing a colon (:) after any flags on a recipe's
first line. This creates a local lockfile based on the destination file name plus whatever has been set
in the LO C K EXT global environment variable.
Alternatively, specify the name of the local lockfile to be used with this recipe after the colon.

1 8 .4 .2 .4 . Spe cial Co ndit io ns and Act io ns
Special characters used before Procmail recipe conditions and actions change the way they are
interpreted.
The following characters may be used after the asterisk character (*) at the beginning of a recipe's
condition line:
! — In the condition line, this character inverts the condition, causing a match to occur only if the
condition does not match the message.
< — Checks if the message is under a specified number of bytes.
> — Checks if the message is over a specified number of bytes.
The following characters are used to perform special actions:
! — In the action line, this character tells Procmail to forward the message to the specified email
addresses.
$ — Refers to a variable set earlier in the rc file. This is often used to set a common mailbox that is
referred to by various recipes.
| — Starts a specified program to process the message.
{ and } — Constructs a nesting block, used to contain additional recipes to apply to matching
messages.
If no special character is used at the beginning of the action line, Procmail assumes that the action
line is specifying the mailbox in which to write the message.

1 8 .4 .2 .5 . Re cipe Exam ple s
Procmail is an extremely flexible program, but as a result of this flexibility, composing Procmail
recipes from scratch can be difficult for new users.
The best way to develop the skills to build Procmail recipe conditions stems from a strong
understanding of regular expressions combined with looking at many examples built by others. A
thorough explanation of regular expressions is beyond the scope of this section. The structure of
Procmail recipes and useful sample Procmail recipes can be found at various places on the Internet.
The proper use and adaptation of regular expressions can be derived by viewing these recipe
examples. In addition, introductory information about basic regular expression rules can be found in
the g rep ( 1) man page.
The following simple examples demonstrate the basic structure of Procmail recipes and can provide
the foundation for more intricate constructions.
A basic recipe may not even contain conditions, as is illustrated in the following example:

4 27

Deployment G uide

:0:
new-mail.spool
The first line specifies that a local lockfile is to be created but does not specify a name, so Procmail
uses the destination file name and appends the value specified in the LO C K EXT environment
variable. No condition is specified, so every message matches this recipe and is placed in the single
spool file called n ew- mail.sp o o l, located within the directory specified by the MAILD IR
environment variable. An MUA can then view messages in this file.
A basic recipe, such as this, can be placed at the end of all rc files to direct messages to a default
location.
The following example matched messages from a specific email address and throws them away.
:0
* ^From: spammer@domain.com
/dev/null
With this example, any messages sent by sp ammer@ d o main .co m are sent to the /d ev/n u ll device,
deleting them.

Sending messages to /dev/null
Be certain that rules are working as intended before sending messages to /d ev/n u ll for
permanent deletion. If a recipe inadvertently catches unintended messages, and those
messages disappear, it becomes difficult to troubleshoot the rule.
A better solution is to point the recipe's action to a special mailbox, which can be checked
from time to time to look for false positives. Once satisfied that no messages are accidentally
being matched, delete the mailbox and direct the action to send the messages to /d ev/n u ll.

The following recipe grabs email sent from a particular mailing list and places it in a specified folder.
:0:
* ^(From|Cc|To).*tux-lug
tuxlug
Any messages sent from the t u x- lu g @ d o main .co m mailing list are placed in the t u xlu g mailbox
automatically for the MUA. Note that the condition in this example matches the message if it has the
mailing list's email address on the Fro m, C c, or T o lines.
Consult the many Procmail online resources available in Section 18.6, “ Additional Resources” for
more detailed and powerful recipes.

1 8 .4 .2 .6 . Spam Filt e rs
Because it is called by Sendmail, Postfix, and Fetchmail upon receiving new emails, Procmail can be
used as a powerful tool for combating spam.
This is particularly true when Procmail is used in conjunction with SpamAssassin. When used
together, these two applications can quickly identify spam emails, and sort or destroy them.

4 28

⁠Chapt er 1 8 . Mail Servers

SpamAssassin uses header analysis, text analysis, blacklists, a spam-tracking database, and selflearning Bayesian spam analysis to quickly and accurately identify and tag spam.

Installing the spamassassin package
In order to use Sp amAssassin , first ensure the spamassassin package is installed on your
system by running, as ro o t :
~]# yu m in st all sp amassassin
For more information on installing packages with Yum, see Section 7.2.4, “ Installing
Packages” .

The easiest way for a local user to use SpamAssassin is to place the following line near the top of the
~ /.p ro cmailrc file:
IN C LU D ER C = /et c/mail/sp amassassin /sp amassassin - d ef au lt .rc
The /et c/mail/sp amassassin /sp amassassin - d ef au lt .rc contains a simple Procmail rule that
activates SpamAssassin for all incoming email. If an email is determined to be spam, it is tagged in
the header as such and the title is prepended with the following pattern:
*****SPAM*****
The message body of the email is also prepended with a running tally of what elements caused it to
be diagnosed as spam.
To file email tagged as spam, a rule similar to the following can be used:
:0 H w * ^X- Sp am- St at u s: Yes sp am
This rule files all email tagged in the header as spam into a mailbox called sp am.
Since SpamAssassin is a Perl script, it may be necessary on busy servers to use the binary
SpamAssassin daemon (sp amd ) and the client application (sp amc). Configuring SpamAssassin
this way, however, requires ro o t access to the host.
To start the sp amd daemon, type the following command:
~]# service sp amassassin st art
To start the SpamAssassin daemon when the system is booted, use an initscript utility, such as the
Services C o n f ig u rat io n T o o l (syst em- co n f ig - services), to turn on the sp amassassin
service. See Chapter 11, Services and Daemons for more information about starting and stopping
services.
To configure Procmail to use the SpamAssassin client application instead of the Perl script, place the
following line near the top of the ~ /.p ro cmailrc file. For a system-wide configuration, place it in
/et c/p ro cmailrc:
IN C LU D ER C = /et c/mail/sp amassassin /sp amassassin - sp amc.rc

4 29

Deployment G uide

18.5. Mail User Agent s
Red Hat Enterprise Linux offers a variety of email programs, both, graphical email client programs,
such as Evo lu t io n , and text-based email programs such as mu t t .
The remainder of this section focuses on securing communication between a client and a server.

18.5.1. Securing Communicat ion
Popular MUAs included with Red Hat Enterprise Linux, such as Evo lu t io n and Mu t t offer SSLencrypted email sessions.
Like any other service that flows over a network unencrypted, important email information, such as
user names, passwords, and entire messages, may be intercepted and viewed by users on the
network. Additionally, since the standard PO P and IMAP protocols pass authentication information
unencrypted, it is possible for an attacker to gain access to user accounts by collecting user names
and passwords as they are passed over the network.

1 8 .5 .1 .1 . Se cure Em ail Clie nt s
Most Linux MUAs designed to check email on remote servers support SSL encryption. To use SSL
when retrieving email, it must be enabled on both the email client and the server.
SSL is easy to enable on the client-side, often done with the click of a button in the MUA's
configuration window or via an option in the MUA's configuration file. Secure IMAP and PO P have
known port numbers (9 9 3 and 9 9 5, respectively) that the MUA uses to authenticate and download
messages.

1 8 .5 .1 .2 . Se curing Em ail Clie nt Co m m unicat io ns
Offering SSL encryption to IMAP and PO P users on the email server is a simple matter.
First, create an SSL certificate. This can be done in two ways: by applying to a Certificate Authority
(CA) for an SSL certificate or by creating a self-signed certificate.

Avoid using self-signed certificates
Self-signed certificates should be used for testing purposes only. Any server used in a
production environment should use an SSL certificate granted by a CA.
To create a self-signed SSL certificate for IMAP or PO P, change to the /et c/p ki/d o veco t / directory,
edit the certificate parameters in the /et c/p ki/d o veco t /d o veco t - o p en ssl.cn f configuration file as
you prefer, and type the following commands, as ro o t :
dovecot]# rm - f cert s/d o veco t .p em p rivat e/d o veco t .p em
dovecot]# /u sr/lib exec/d o veco t /mkcert .sh
Once finished, make sure you have the following configurations in your /et c/d o veco t /co n f .d /10ssl.co n f file:
ssl_cert =  prompt, you have successfully logged in. Once you are logged in, type
h elp for a list of commands. If you want to browse the contents of your home directory,
replace sharename with your user name. If the - U switch is not used, the user name of the
current user is passed to the Samba server.
3. To exit smb clien t , type exit at the smb :\> prompt.

2 0 .1 .3.1 . Mo unt ing t he Share
Sometimes it is useful to mount a Samba share to a directory so that the files in the directory can be
treated as if they are part of the local file system.
To mount a Samba share to a directory, create a directory to mount it to (if it does not already exist),
and execute the following command as ro o t :
mo u n t - t cif s //servername/sharename /mnt/point/ - o
u sern ame= username,p asswo rd = password
This command mounts sharename from servername in the local directory /mnt/point/.
For more information about mounting a samba share, see the mount.cifs(8) manual page.

Installing cifs-utils package
The mo u n t .cif s utility is a separate RPM (independent from Samba). In order to use
mo u n t .cif s, first ensure the cifs-utils package is installed on your system by running the
following command as ro o t :
~]# yu m in st all cif s- u t ils
For more information on installing packages with Yum, see Section 7.2.4, “ Installing
Packages” .
Note that the cifs-utils package also contains the cif s.u p call binary called by the kernel in
order to perform kerberized CIFS mounts. For more information on cif s.u p call, see the
cifs.upcall(8) manual page.

CIFS servers that require plain text passwords
Some CIFS servers require plain text passwords for authentication. Support for plain text
password authentication can be enabled using the following command as ro o t :
~]# ech o 0x37 > /p ro c/f s/cif s/Secu rit yFlag s
WARNING: This operation can expose passwords by removing password encryption.

20.1.4 . Configuring a Samba Server

4 51

Deployment G uide

The default configuration file (/et c/samb a/smb .co n f ) allows users to view their home directories as
a Samba share. It also shares all printers configured for the system as Samba shared printers. You
can attach a printer to the system and print to it from the Windows machines on your network.

2 0 .1 .4 .1 . Graphical Co nfigurat io n
To configure Samba using a graphical interface, use one of the available Samba graphical user
interfaces. A list of available GUIs can be found at http://www.samba.org/samba/GUI/.

2 0 .1 .4 .2 . Co m m and-Line Co nfigurat io n
Samba uses /et c/samb a/smb .co n f as its configuration file. If you change this configuration file, the
changes do not take effect until you restart the Samba daemon with the following command as ro o t :
~]# service smb rest art
To specify the Windows workgroup and a brief description of the Samba server, edit the following
lines in your /et c/samb a/smb .co n f file:
workgroup = WORKGROUPNAME
server string = BRIEF COMMENT ABOUT SERVER
Replace WORKGROUPNAME with the name of the Windows workgroup to which this machine should
belong. The BRIEF COMMENT ABOUT SERVER is optional and is used as the Windows comment
about the Samba system.
To create a Samba share directory on your Linux system, add the following section to your
/et c/samb a/smb .co n f file (after modifying it to reflect your needs and your system):

Examp le 20.1. An Examp le C o n f ig u rat io n o f a Samb a Server
[sharename]
comment = Insert a comment here
path = /home/share/
valid users = tfox carole
writable = yes
create mask = 0765

The above example allows the users t f o x and caro le to read and write to the directory
/h o me/sh are/, on the Samba server, from a Samba client.

2 0 .1 .4 .3. Encrypt e d Passwo rds
Encrypted passwords are enabled by default because it is more secure to use them. To create a user
with an encrypted password, use the smb p asswd utility:
smb p asswd - a username

20.1.5. St art ing and St opping Samba
To start a Samba server, type the following command in a shell prompt, as ro o t :

4 52

⁠Chapt er 2 0 . File and Print Servers

~]# service smb st art

Setting up a domain member server
To set up a domain member server, you must first join the domain or Active D irectory using the
n et jo in command before starting the smb service. Also it is recommended to run win b in d
before smb d .
To stop the server, type the following command in a shell prompt, as ro o t :
~]# service smb st o p
The rest art option is a quick way of stopping and then starting Samba. This is the most reliable way
to make configuration changes take effect after editing the configuration file for Samba. Note that the
restart option starts the daemon even if it was not running originally.
To restart the server, type the following command in a shell prompt, as ro o t :
~]# service smb rest art
The co n d rest art (conditional restart) option only starts smb on the condition that it is currently
running. This option is useful for scripts, because it does not start the daemon if it is not running.

Applying the changes to the configuration
When the /et c/samb a/smb .co n f file is changed, Samba automatically reloads it after a few
minutes. Issuing a manual rest art or relo ad is just as effective.
To conditionally restart the server, type the following command as ro o t :
~]# service smb co n d rest art
A manual reload of the /et c/samb a/smb .co n f file can be useful in case of a failed automatic reload
by the smb service. To ensure that the Samba server configuration file is reloaded without restarting
the service, type the following command, as ro o t :
~]# service smb relo ad
By default, the smb service does not start automatically at boot time. To configure Samba to start at
boot time, use an initscript utility, such as /sb in /ch kco n f ig , /u sr/sb in /n t sysv, or the Services
C o n f ig u rat io n T o o l program. See Chapter 11, Services and Daemons for more information
regarding these tools.

20.1.6. Samba Server T ypes and t he smb .co n f File
Samba configuration is straightforward. All modifications to Samba are done in the
/et c/samb a/smb .co n f configuration file. Although the default smb .co n f file is well documented, it
does not address complex topics such as LD AP, Active D irectory, and the numerous domain
controller implementations.

4 53

Deployment G uide

The following sections describe the different ways a Samba server can be configured. Keep in mind
your needs and the changes required to the /et c/samb a/smb .co n f file for a successful
configuration.

2 0 .1 .6 .1 . St and-alo ne Se rve r
A stand-alone server can be a workgroup server or a member of a workgroup environment. A standalone server is not a domain controller and does not participate in a domain in any way. The
following examples include several user-level security configurations. For more information on
security modes, see Section 20.1.7, “ Samba Security Modes” .

Ano nym o us Re ad-Only
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement
anonymous read-only file sharing. Two directives are used to configure anonymous access – map to
guest = Bad user and guest account = nobody.

Examp le 20.2. An Examp le C o n f ig u rat io n o f a An o n ymo u s R ead - O n ly Samb a Server
[global]
workgroup = D OCS
netbios name = D OCS_SRV
security = user
guest account = nobody # default value
map to guest = Bad user
[data]
comment = D ocumentation Samba Server
path = /export
read only = yes
guest ok = yes

Ano nym o us Re ad/Writ e
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement
anonymous read/write file sharing. To enable anonymous read/write file sharing, set the read only
directive to n o . The force user and force group directives are also added to enforce the ownership of
any newly placed files specified in the share.

Do not use anonymous read/write servers
Although having an anonymous read/write server is possible, it is not recommended. Any files
placed in the share space, regardless of user, are assigned the user/group combination as
specified by a generic user (force user) and group (force group) in the /et c/samb a/smb .co n f
file.

Examp le 20.3. An Examp le C o n f ig u rat io n o f a An o n ymo u s R ead /Writ e Samb a Server
[global]
workgroup = D OCS

4 54

⁠Chapt er 2 0 . File and Print Servers

security = user
guest account = nobody # default value
map to guest = Bad user
[data]
comment = D ata
path = /export
guest ok = yes
writeable = yes
force user = user
force group = group

Ano nym o us Print Se rve r
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement an
anonymous print server. Setting browseable to n o as shown does not list the printer in Windows
N et wo rk N eig h b o rh o o d . Although hidden from browsing, configuring the printer explicitly is
possible. By connecting to D O C S_SR V using NetBIOS, the client can have access to the printer if
the client is also part of the D O C S workgroup. It is also assumed that the client has the correct local
printer driver installed, as the use client driver directive is set to yes. In this case, the Samba server
has no responsibility for sharing printer drivers to the client.

Examp le 20.4 . An Examp le C o n f ig u rat io n o f a An o n ymo u s Prin t Samb a Server
[global]
workgroup = D OCS
netbios name = D OCS_SRV
security = user
map to guest = Bad user
printing = cups
[printers]
comment = All Printers
path = /var/spool/samba
guest ok = yes
printable = yes
use client driver = yes
browseable = yes

Se cure Re ad/Writ e File and Print Se rve r
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement a
secure read/write file and print server. Setting the security directive to u ser forces Samba to
authenticate client connections. Notice the [ h o mes] share does not have a force user or force group
directive as the [ p u b lic] share does. The [ h o mes] share uses the authenticated user details for any
files created as opposed to the force user and force group in [ p u b lic] .

Examp le 20.5. An Examp le C o n f ig u rat io n o f a Secu re R ead /Writ e File an d Prin t Samb a
Server
[global]

4 55

Deployment G uide

workgroup = D OCS
netbios name = D OCS_SRV
security = user
printcap name = cups
disable spools = yes
show add printer wizard = no
printing = cups
[homes]
comment = Home D irectories
valid users = % S
read only = no
browseable = no
[public]
comment = D ata
path = /export
force user = docsbot
force group = users
guest ok = yes
[printers]
comment = All Printers
path = /var/spool/samba
printer admin = john, ed, @admins
create mask = 0600
guest ok = yes
printable = yes
use client driver = yes
browseable = yes

2 0 .1 .6 .2 . Do m ain Me m be r Se rve r
A domain member, while similar to a stand-alone server, is logged into a domain controller (either
Windows or Samba) and is subject to the domain's security rules. An example of a domain member
server would be a departmental server running Samba that has a machine account on the Primary
D omain Controller (PD C). All of the department's clients still authenticate with the PD C, and desktop
profiles and all network policy files are included. The difference is that the departmental server has
the ability to control printer and network shares.

Act ive Dire ct o ry Do m ain Me m be r Se rve r
To implement an Active D irectory domain member server, follow procedure below:
Pro ced u re 20.3. Ad d in g a Memb er Server t o an Act ive D irect o ry D o main
1. Create the /et c/samb a/smb .co n f configuration file on a member server to be added to the
Active D irectory domain. Add the following lines to the configuration file:
[global]
realm = EXAMPLE.COM
security = AD S
encrypt passwords = yes

4 56

⁠Chapt er 2 0 . File and Print Servers

# Optional. Use only if Samba cannot determine the Kerberos server automatically.
password server = kerberos.example.com
With the above configuration, Samba authenticates users for services being run locally but is
also a client of the Active D irectory. Ensure that your kerberos realm parameter is shown in all
caps (for example realm = EXAMPLE.C O M). Since Windows 2000/2003/2008 requires
Kerberos for Active D irectory authentication, the realm directive is required. If Active D irectory
and Kerberos are running on different servers, the password server directive is required to
help the distinction.
2. Configure Kerberos on the member server. Create the /et c/krb 5.co n f configuration file with
the following content:
[logging]
default = FILE:/var/log/krb5libs.log
[libdefaults]
default_realm = AD .EXAMPLE.COM
dns_lookup_realm = true
dns_lookup_kdc = true
ticket_lifetime = 24h
renew_lifetime = 7d
rdns = false
forwardable = false
[realms]
# D efine only if D NS lookups are not working
# AD .EXAMPLE.COM = {
# kdc = server.ad.example.com
# admin_server = server.ad.example.com
# master_kdc = server.ad.example.com
#}
[domain_realm]
# D efine only if D NS lookups are not working
# .ad.example.com = AD .EXAMPLE.COM
# ad.example.com = AD .EXAMPLE.COM
Uncomment the [ realms] and [ d o main _realm] sections if D NS lookups are not working.
For more information on Kerberos, and the /et c/krb 5.co n f file, see the Using Kerberos section
of the Red Hat Enterprise Linux 6 Managing Single Sign-On and Smart Cards.
3. To join an Active D irectory server, type the following command as ro o t on the member server:
~]# n et ad s jo in - U ad min ist rat o r% password
The n et command authenticates as Ad min ist rat o r using the NT LAN Manager (NTLM)
protocol and creates the machine account. Then n et uses the machine account credentials
to authenticate with Kerberos.

4 57

Deployment G uide

T he security option
Since security = ads and not security = user is used, a local password back end such
as smb p asswd is not needed. Older clients that do not support security = ads are
authenticated as if security = domain had been set. This change does not affect
functionality and allows local users not previously in the domain.

Windo ws NT 4 -base d Do m ain Me m be r Se rve r
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement a
Windows NT4-based domain member server. Becoming a member server of an NT4-based domain is
similar to connecting to an Active D irectory. The main difference is NT4-based domains do not use
Kerberos in their authentication method, making the /et c/samb a/smb .co n f file simpler. In this
instance, the Samba member server functions as a pass through to the NT4-based domain server.

Examp le 20.6 . An Examp le C o n f ig u rat io n o f Samb a Win d o ws N T 4 - b ased D o main
Memb er Server
[global]
workgroup = D OCS
netbios name = D OCS_SRV
security = domain
[homes]
comment = Home D irectories
valid users = % S
read only = no
browseable = no
[public]
comment = D ata
path = /export
force user = docsbot
force group = users
guest ok = yes

Having Samba as a domain member server can be useful in many situations. There are times where
the Samba server can have other uses besides file and printer sharing. It may be beneficial to make
Samba a domain member server in instances where Linux-only applications are required for use in
the domain environment. Administrators appreciate keeping track of all machines in the domain, even
if not Windows-based. In the event the Windows-based server hardware is deprecated, it is quite easy
to modify the /et c/samb a/smb .co n f file to convert the server to a Samba-based PD C. If Windows
NT-based servers are upgraded to Windows 2000/2003/2008 the /et c/samb a/smb .co n f file is
easily modifiable to incorporate the infrastructure change to Active D irectory if needed.

4 58

⁠Chapt er 2 0 . File and Print Servers

Make sure you join the domain before starting Samba
After configuring the /et c/samb a/smb .co n f file, join the domain before starting Samba by
typing the following command as ro o t :
~]# n et rp c jo in - U ad min ist rat o r% p asswo rd

Note that the - S option, which specifies the domain server host name, does not need to be stated in
the n et rp c jo in command. Samba uses the host name specified by the workgroup directive in the
/et c/samb a/smb .co n f file instead of it being stated explicitly.

2 0 .1 .6 .3. Do m ain Co nt ro lle r
A domain controller in Windows NT is functionally similar to a Network Information Service (NIS)
server in a Linux environment. D omain controllers and NIS servers both host user and group
information databases as well as related services. D omain controllers are mainly used for security,
including the authentication of users accessing domain resources. The service that maintains the
user and group database integrity is called the Security Account Manager (SAM). The SAM database is
stored differently between Windows and Linux Samba-based systems, therefore SAM replication
cannot be achieved and platforms cannot be mixed in a PD C/BD C environment.
In a Samba environment, there can be only one PD C and zero or more BD Cs.

A mixed Samba/Windows domain controller environment
Samba cannot exist in a mixed Samba/Windows domain controller environment (Samba
cannot be a BD C of a Windows PD C or vice versa). Alternatively, Samba PD Cs and BD Cs can
coexist.

Prim ary Do m ain Co nt ro lle r (PDC) Using t d b sam
The simplest and most common implementation of a Samba PD C uses the new default t d b sam
password database back end. Replacing the aging smb p asswd back end, t d b sam has numerous
improvements that are explained in more detail in Section 20.1.8, “ Samba Account Information
D atabases” . The passdb backend directive controls which back end is to be used for the PD C.
The following /et c/samb a/smb .co n f file shows a sample configuration needed to implement a
t d b sam password database back end.

Examp le 20.7. An Examp le C o n f ig u rat io n o f Primary D o main C o n t ro ller ( PD C ) U sin g
t d b sam
[global]
workgroup = D OCS
netbios name = D OCS_SRV
passdb backend = tdbsam
security = user
add user script = /usr/sbin/useradd -m " % u"
delete user script = /usr/sbin/userdel -r " % u"

4 59

Deployment G uide

add group script = /usr/sbin/groupadd " % g"
delete group script = /usr/sbin/groupdel " % g"
add user to group script = /usr/sbin/usermod -G " % g" " % u"
add machine script = /usr/sbin/useradd -s /bin/false -d /dev/null -g machines " % u"
# The following specifies the default logon script
# Per user logon scripts can be specified in the user
# account using pdbedit logon script = logon.bat
# This sets the default profile path.
# Set per user paths with pdbedit
logon drive = H:
domain logons = yes
os level = 35
preferred master = yes
domain master = yes
[homes]
comment = Home D irectories
valid users = % S
read only = no
[netlogon]
comment = Network Logon Service
path = /var/lib/samba/netlogon/scripts
browseable = no
read only = no
# For profiles to work, create a user directory under the
# path shown.
# mkdir -p /var/lib/samba/profiles/john
[Profiles]
comment = Roaming Profile Share
path = /var/lib/samba/profiles
read only = no
browseable = no
guest ok = yes
profile acls = yes
# Other resource shares ... ...

To provide a functional PD C system which uses t d b sam follow these steps:
1. Adjust the smb .co n f configuration file as shown in Example 20.7, “ An Example
Configuration of Primary D omain Controller (PD C) Using t d b sam” .
2. Add the ro o t user to the Samba password database. You will be prompted to provide a new
Samba password for the ro o t user:
~]# smb p asswd - a ro o t
New SMB password:
3. Start the smb service:
~]# service smb st art
4. Make sure all profile, user, and netlogon directories are created.

4 60

⁠Chapt er 2 0 . File and Print Servers

5. Add groups that users can be members of:
~]# g ro u p ad d - f u sers
~]# g ro u p ad d - f n o b o d y
~]# g ro u p ad d - f n t ad min s
6. Associate the UNIX groups with their respective Windows groups.
~]# n et g ro u p map ad d n t g ro u p = "D o main U sers" u n ixg ro u p = u sers
~]# n et g ro u p map ad d n t g ro u p = "D o main G u est s" u n ixg ro u p = n o b o d y
~]# n et g ro u p map ad d n t g ro u p = "D o main Ad min s" u n ixg ro u p = n t ad min s
7. Grant access rights to a user or a group. For example, to grant the right to add client
machines to the domain on a Samba domain controller, to the members to the D omain
Admins group, execute the following command:
~]# n et rp c rig h t s g ran t ' D O C S\D o main Ad min s' Set Mach in eAcco u n t Privileg e S PD C - U ro o t
Keep in mind that Windows systems prefer to have a primary group which is mapped to a domain
group such as D omain Users.
Windows groups and users use the same namespace thus not allowing the existence of a group and
a user with the same name like in UNIX.

Limitations of the tdbsam authentication back end
If you need more than one domain controller or have more than 250 users, do not use the
t d b sam authentication back end. LD AP is recommended in these cases.

Prim ary Do m ain Co nt ro lle r (PDC) wit h Act ive Dire ct o ry
Although it is possible for Samba to be a member of an Active D irectory, it is not possible for Samba
to operate as an Active D irectory domain controller.

20.1.7. Samba Securit y Modes
There are only two types of security modes for Samba, share-level and user-level, which are
collectively known as security levels. Share-level security is deprecated and Red Hat recommends to
use user-level security instead. User-level security can be implemented in one of three different ways.
The different ways of implementing a security level are called security modes.

2 0 .1 .7 .1 . Use r-Le ve l Se curit y
User-level security is the default and recommended setting for Samba. Even if the security = user
directive is not listed in the /et c/samb a/smb .co n f file, it is used by Samba. If the server accepts the
client's user name and password, the client can then mount multiple shares without specifying a
password for each instance. Samba can also accept session-based user name and password
requests. The client maintains multiple authentication contexts by using a unique UID for each
logon.
In the /et c/samb a/smb .co n f file, the security = user directive that sets user-level security is:

4 61

Deployment G uide

[GLOBAL]
...
security = user
...

Sam ba Gue st Share s
As mentioned above, share-level security mode is deprecated and highly recommended to not use.
To configure a Samba guest share without using the security = share parameter, follow the procedure
below:
Pro ced u re 20.4 . C o n f ig u rin g Samb a G u est Sh ares
1. Create a username map file, in this example /et c/samb a/smb u sers, and add the following
line to it:
nobody = guest
2. Add the following directives to the main section in the /et c/samb a/smb .co n f file. Also, do
not use the valid users directive:
[GLOBAL]
...
security = user
map to guest = Bad User
username map = /etc/samba/smbusers
...
The username map directive provides a path to the username map file specified in the
previous step.
3. Add the following directive to the share section in the /ect /samb a/smb .co n f file. D o not use
the valid users directive.
[SHARE]
...
guest ok = yes
...
The following sections describe other implementations of user-level security.

Do m ain Se curit y Mo de (Use r-Le ve l Se curit y)
In domain security mode, the Samba server has a machine account (domain security trust account)
and causes all authentication requests to be passed through to the domain controllers. The Samba
server is made into a domain member server by using the following directives in the
/et c/samb a/smb .co n f file:
[GLOBAL]
...
security = domain
workgroup = MARKETING
...

4 62

⁠Chapt er 2 0 . File and Print Servers

Act ive Dire ct o ry Se curit y Mo de (Use r-Le ve l Se curit y)
If you have an Active D irectory environment, it is possible to join the domain as a native Active
D irectory member. Even if a security policy restricts the use of NT-compatible authentication
protocols, the Samba server can join an AD S using Kerberos. Samba in Active D irectory member
mode can accept Kerberos tickets.
In the /et c/samb a/smb .co n f file, the following directives make Samba an Active D irectory member
server:
[GLOBAL]
...
security = AD S
realm = EXAMPLE.COM
password server = kerberos.example.com
...

2 0 .1 .7 .2 . Share -Le ve l Se curit y
With share-level security, the server accepts only a password without an explicit user name from the
client. The server expects a password for each share, independent of the user name. There have
been recent reports that Microsoft Windows clients have compatibility issues with share-level security
servers. This mode is deprecated and Red Hat strongly discourage use of share-level security.
Follow steps in Procedure 20.4, “ Configuring Samba Guest Shares” instead of using the security =
share directive.

20.1.8. Samba Account Informat ion Dat abases
The following is a list different back ends you can use with Samba. Other back ends not listed here
may also be available.
Plain T ext
Plain text back ends are nothing more than the /et c/p asswd type back ends. With a plain
text back end, all user names and passwords are sent unencrypted between the client and
the Samba server. This method is very insecure and is not recommended for use by any
means. It is possible that different Windows clients connecting to the Samba server with
plain text passwords cannot support such an authentication method.
smb p asswd
The smb p asswd back end utilizes a plain ASCII text layout that includes the MS Windows
LanMan and NT account, and encrypted password information. The smb p asswd back
end lacks the storage of the Windows NT/2000/2003 SAM extended controls. The
smb p asswd back end is not recommended because it does not scale well or hold any
Windows information, such as RID s for NT-based groups. The t d b sam back end solves
these issues for use in a smaller database (250 users), but is still not an enterprise-class
solution.
ld ap sam_co mp at
The ld ap sam_co mp at back end allows continued OpenLD AP support for use with
upgraded versions of Samba.
t d b sam
The default t d b sam password back end provides a database back end for local servers,

4 63

Deployment G uide

servers that do not need built-in database replication, and servers that do not require the
scalability or complexity of LD AP. The t d b sam back end includes all of the smb p asswd
database information as well as the previously-excluded SAM information. The inclusion of
the extended SAM data allows Samba to implement the same account and system access
controls as seen with Windows NT/2000/2003/2008-based systems.
The t d b sam back end is recommended for 250 users at most. Larger organizations should
require Active D irectory or LD AP integration due to scalability and possible network
infrastructure concerns.
ld ap sam
The ld ap sam back end provides an optimal distributed account installation method for
Samba. LD AP is optimal because of its ability to replicate its database to any number of
servers such as the R ed H at D irect o ry Server or an O p en LD AP Server. LD AP
databases are light-weight and scalable, and as such are preferred by large enterprises.
Installation and configuration of directory servers is beyond the scope of this chapter. For
more information on the R ed H at D irect o ry Server, see the Red Hat Directory Server 9.0
Deployment Guide. For more information on LD AP, see Section 19.1, “ OpenLD AP” .
If you are upgrading from a previous version of Samba to 3.0, note that the OpenLD AP
schema file (/u sr/sh are/d o c/samb a- version/LD AP/samb a.sch ema) and the Red Hat
D irectory Server schema file (/u sr/sh are/d o c/samb a- version/LD AP/samb a- sch emaFD S.ld if ) have changed. These files contain the attribute syntax definitions and objectclass
definitions that the ld ap sam back end needs in order to function properly.
As such, if you are using the ld ap sam back end for your Samba server, you will need to
configure slap d to include one of these schema file. See Section 19.1.3.3, “ Extending
Schema” for directions on how to do this.

Make sure the openldap-servers package is installed
You need to have the openldap-servers package installed if you want to use the
ld ap sam back end. To ensure that the package is installed, execute the following
command as ro o t s:
~]# yu m in st all o p en ld ap - servers

20.1.9. Samba Net work Browsing
Network browsing enables Windows and Samba servers to appear in the Windows N et wo rk
N eig h b o rh o o d . Inside the N et wo rk N eig h b o rh o o d , icons are represented as servers and if
opened, the server's shares and printers that are available are displayed.
Network browsing capabilities require NetBIOS over T C P/IP. NetBIOS-based networking uses
broadcast (U D P) messaging to accomplish browse list management. Without NetBIOS and WINS as
the primary method for T C P/IP host name resolution, other methods such as static files (/et c/h o st s)
or D N S, must be used.
A domain master browser collates the browse lists from local master browsers on all subnets so that
browsing can occur between workgroups and subnets. Also, the domain master browser should
preferably be the local master browser for its own subnet.

4 64

⁠Chapt er 2 0 . File and Print Servers

2 0 .1 .9 .1 . Do m ain Bro wsing
By default, a Windows server PD C for a domain is also the domain master browser for that domain. A
Samba server must not be set up as a domain master server in this type of situation.
For subnets that do not include the Windows server PD C, a Samba server can be implemented as a
local master browser. Configuring the /et c/samb a/smb .co n f file for a local master browser (or no
browsing at all) in a domain controller environment is the same as workgroup configuration (see
Section 20.1.4, “ Configuring a Samba Server” ).

2 0 .1 .9 .2 . WINS (Windo ws Int e rne t Nam e Se rve r)
Either a Samba server or a Windows NT server can function as a WINS server. When a WINS server is
used with NetBIOS enabled, UD P unicasts can be routed which allows name resolution across
networks. Without a WINS server, the UD P broadcast is limited to the local subnet and therefore
cannot be routed to other subnets, workgroups, or domains. If WINS replication is necessary, do not
use Samba as your primary WINS server, as Samba does not currently support WINS replication.
In a mixed NT/2000/2003/2008 server and Samba environment, it is recommended that you use the
Microsoft WINS capabilities. In a Samba-only environment, it is recommended that you use only one
Samba server for WINS.
The following is an example of the /et c/samb a/smb .co n f file in which the Samba server is serving
as a WINS server:

Examp le 20.8. An Examp le C o n f ig u rat io n o f WIN S Server
[global]
wins support = yes

Using WINS
All servers (including Samba) should connect to a WINS server to resolve NetBIOS names.
Without WINS, browsing only occurs on the local subnet. Furthermore, even if a domain-wide
list is somehow obtained, hosts cannot be resolved for the client without WINS.

20.1.10. Samba wit h CUPS Print ing Support
Samba allows client machines to share printers connected to the Samba server. In addition, Samba
also allows client machines to send documents built in Linux to Windows printer shares. Although
there are other printing systems that function with Red Hat Enterprise Linux, CUPS (Common UNIX
Print System) is the recommended printing system due to its close integration with Samba.

2 0 .1 .1 0 .1 . Sim ple smb .co n f Se t t ings
The following example shows a very basic /et c/samb a/smb .co n f configuration for CUPS support:

Examp le 20.9 . An Examp le C o n f ig u rat io n o f Samb a wit h C U PS Su p p o rt
[global]

4 65

Deployment G uide

load printers = yes
printing = cups
printcap name = cups
[printers]
comment = All Printers
path = /var/spool/samba
browseable = no
guest ok = yes
writable = no
printable = yes
printer admin = @ntadmins
[print$]
comment = Printer D rivers Share
path = /var/lib/samba/drivers
write list = ed, john
printer admin = ed, john

Other printing configurations are also possible. To add additional security and privacy for printing
confidential documents, users can have their own print spooler not located in a public path. If a job
fails, other users would not have access to the file.
The print$ directive contains printer drivers for clients to access if not available locally. The print$
directive is optional and may not be required depending on the organization.
Setting browseable to yes enables the printer to be viewed in the Windows Network Neighborhood,
provided the Samba server is set up correctly in the domain or workgroup.

20.1.11. Samba Dist ribut ion Programs
f in d smb
f in d smb 
The f in d smb program is a Perl script which reports information about SMB -aware systems on a
specific subnet. If no subnet is specified the local subnet is used. Items displayed include IP address,
NetBIOS name, workgroup or domain name, operating system, and version. The f in d smb command
is used in the following format:
The following example shows the output of executing f in d smb as any valid user on a system:
~]$ f in d smb
IP AD D R
NETBIOS NAME WORKGROUP/OS/VERSION
-----------------------------------------------------------------10.1.59.25 VERVE
[MYGROUP] [Unix] [Samba 3.0.0-15]
10.1.59.26 STATION22 [MYGROUP] [Unix] [Samba 3.0.2-7.FC1]
10.1.56.45 TREK
+[WORKGROUP] [Windows 5.0] [Windows 2000 LAN Manager]
10.1.57.94 PIXEL
[MYGROUP] [Unix] [Samba 3.0.0-15]
10.1.57.137 MOBILE001 [WORKGROUP] [Windows 5.0] [Windows 2000 LAN Manager]
10.1.57.141 JAWS
+[KWIKIMART] [Unix] [Samba 2.2.7a-security-rollup-fix]
10.1.56.159 FRED
+[MYGROUP] [Unix] [Samba 3.0.0-14.3E]
10.1.59.192 LEGION
*[MYGROUP] [Unix] [Samba 2.2.7-security-rollup-fix]
10.1.56.205 NANCYN
+[MYGROUP] [Unix] [Samba 2.2.7a-security-rollup-fix]
n et

4 66

⁠Chapt er 2 0 . File and Print Servers

n et    
The n et utility is similar to the n et utility used for Windows and MS-D OS. The first argument is used
to specify the protocol to use when executing a command. The protocol option can be ad s, rap , or
rp c for specifying the type of server connection. Active D irectory uses ad s, Win9x/NT3 uses rap , and
Windows NT4/2000/2003/2008 uses rp c. If the protocol is omitted, n et automatically tries to
determine it.
The following example displays a list of the available shares for a host named wakko :
~]$ n et - l sh are - S wakko
Password:
Enumerating shared resources (exports) on remote server:
Share name Type D escription
---------- ---- ----------data
D isk Wakko data share
tmp
D isk Wakko tmp share
IPC$
IPC
IPC Service (Samba Server)
AD MIN$
IPC
IPC Service (Samba Server)
The following example displays a list of Samba users for a host named wakko :
~]$ n et - l u ser - S wakko
root password:
User name
Comment
----------------------------andriusb
D ocumentation
joe
Marketing
lisa
Sales
n mb lo o ku p
n mb lo o ku p  
The n mb lo o ku p program resolves NetBIOS names into IP addresses. The program broadcasts its
query on the local subnet until the target machine replies.
The following example displays the IP address of the NetBIOS name t rek:
~]$ n mb lo o ku p t rek
querying trek on 10.1.59.255
10.1.56.45 trek<00>
p d b ed it
p d b ed it 
The p d b ed it program manages accounts located in the SAM database. All back ends are supported
including smb p asswd , LD AP, and the tdb database library.
The following are examples of adding, deleting, and listing users:

4 67

Deployment G uide

~]$ p d b ed it - a krist in
new password:
retype new password:
Unix username:
kristin
NT username:
Account Flags:
[U
]
User SID :
S-1-5-21-1210235352-3804200048-1474496110-2012
Primary Group SID : S-1-5-21-1210235352-3804200048-1474496110-2077
Full Name: Home D irectory:
\\wakko\kristin
HomeD ir D rive:
Logon Script:
Profile Path:
\\wakko\kristin\profile
D omain:
WAKKO
Account desc:
Workstations: Munged
dial:
Logon time:
0
Logoff time:
Mon, 18 Jan 2038 22:14:07 GMT
Kickoff time:
Mon, 18 Jan 2038 22:14:07 GMT
Password last set: Thu, 29 Jan 2004 08:29:28
GMT Password can change: Thu, 29 Jan 2004 08:29:28 GMT
Password must change: Mon, 18 Jan 2038 22:14:07 GMT
~]$ p d b ed it - v - L krist in
Unix username:
kristin
NT username:
Account Flags:
[U
]
User SID :
S-1-5-21-1210235352-3804200048-1474496110-2012
Primary Group SID : S-1-5-21-1210235352-3804200048-1474496110-2077
Full Name:
Home D irectory:
\\wakko\kristin
HomeD ir D rive:
Logon Script:
Profile Path:
\\wakko\kristin\profile
D omain:
WAKKO
Account desc:
Workstations: Munged
dial:
Logon time:
0
Logoff time:
Mon, 18 Jan 2038 22:14:07 GMT
Kickoff time:
Mon, 18 Jan 2038 22:14:07 GMT
Password last set: Thu, 29 Jan 2004 08:29:28 GMT
Password can change: Thu, 29 Jan 2004 08:29:28 GMT
Password must change: Mon, 18 Jan 2038 22:14:07 GMT
~]$ p d b ed it - L
andriusb:505:
joe:503:
lisa:504:
kristin:506:
~]$ p d b ed it - x jo e
~]$ p d b ed it - L
andriusb:505: lisa:504: kristin:506:
rp cclien t
rp cclien t  

4 68

⁠Chapt er 2 0 . File and Print Servers

The rp cclien t program issues administrative commands using Microsoft RPCs, which provide
access to the Windows administration graphical user interfaces (GUIs) for systems management.
This is most often used by advanced users that understand the full complexity of Microsoft RPCs.
smb cacls
smb cacls   
The smb cacls program modifies Windows ACLs on files and directories shared by a Samba server
or a Windows server.
smb clien t
smb clien t   
The smb clien t program is a versatile UNIX client which provides functionality similar to the f t p
utility.
smb co n t ro l
smb co n t ro l - i 
smb co n t ro l    
The smb co n t ro l program sends control messages to running smb d , n mb d , or win b in d d
daemons. Executing smb co n t ro l - i runs commands interactively until a blank line or a 'q' is
entered.
smb p asswd
smb p asswd   
The smb p asswd program manages encrypted passwords. This program can be run by a superuser
to change any user's password and also by an ordinary user to change their own Samba password.
smb sp o o l
smb sp o o l